Nothing
R/bayes.anova.R
In bayesanova: Bayesian Inference in the Analysis of Variance via Markov Chain Monte Carlo in Gaussian Mixture Models
Defines functions
post.pred.check
assumption.check
anovaplot
bayes.anova
Documented in
anovaplot
assumption.check
bayes.anova
post.pred.check
bayes.anova <- function(n=10000,first,second,third,fourth=NULL,fifth=NULL,sixth=NULL,hyperpars="custom",burnin=n/2,sd="sd",q=0.1,ci=0.95){
# utility function for computing the mode
getmode <- function(v) {
uniqv <- unique(v)
uniqv[which.max(tabulate(match(v, uniqv)))]
}
Nsim=n # number of steps performed by the Gibbs sampler
if(length(first)>1){
firstComponent = first # set first component
} else {
warning("Please provide at least data for three groups. Data for first group is missing.")
}
if(length(second)>1){
secondComponent = second # set second component
} else {
warning("Please provide at least data for three groups. Data for second group is missing.")
}
if(length(third)>1){
thirdComponent = third # set second component
} else {
warning("Please provide at least data for three groups. Data for third group is missing.")
}
if(length(fourth)>1){
fourthComponent = fourth # set second component
}
if(length(fifth)>1){
fifthComponent = fifth # set second component
}
if(length(sixth)>1){
sixthComponent = sixth # set second component
}
mu1 <- numeric(Nsim) # arrays for parameters in first chain
mu2 <- numeric(Nsim)
mu3 <- numeric(Nsim)
if(length(fourth)>1){
mu4 <- numeric(Nsim)
}
if(length(fifth)>1){
mu5 <- numeric(Nsim)
}
if(length(sixth)>1){
mu6 <- numeric(Nsim)
}
sigma1Sq <- numeric(Nsim)
sigma2Sq <- numeric(Nsim)
sigma3Sq <- numeric(Nsim)
if(length(fourth)>1){
sigma4Sq <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5Sq <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6Sq <- numeric(Nsim)
}
mu1SecC <- numeric(Nsim) # arrays for parameters in first chain
mu2SecC <- numeric(Nsim)
mu3SecC <- numeric(Nsim)
if(length(fourth)>1){
mu4SecC <- numeric(Nsim)
}
if(length(fifth)>1){
mu5SecC <- numeric(Nsim)
}
if(length(sixth)>1){
mu6SecC <- numeric(Nsim)
}
sigma1SqSecC <- numeric(Nsim)
sigma2SqSecC <- numeric(Nsim)
sigma3SqSecC <- numeric(Nsim)
if(length(fourth)>1){
sigma4SqSecC <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5SqSecC <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6SqSecC <- numeric(Nsim)
}
mu1ThdC <- numeric(Nsim) # arrays for parameters in first chain
mu2ThdC <- numeric(Nsim)
mu3ThdC <- numeric(Nsim)
if(length(fourth)>1){
mu4ThdC <- numeric(Nsim)
}
if(length(fifth)>1){
mu5ThdC <- numeric(Nsim)
}
if(length(sixth)>1){
mu6ThdC <- numeric(Nsim)
}
sigma1SqThdC <- numeric(Nsim)
sigma2SqThdC <- numeric(Nsim)
sigma3SqThdC <- numeric(Nsim)
if(length(fourth)>1){
sigma4SqThdC <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5SqThdC <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6SqThdC <- numeric(Nsim)
}
mu1FrtC <- numeric(Nsim) # arrays for parameters in first chain
mu2FrtC <- numeric(Nsim)
mu3FrtC <- numeric(Nsim)
if(length(fourth)>1){
mu4FrtC <- numeric(Nsim)
}
if(length(fifth)>1){
mu5FrtC <- numeric(Nsim)
}
if(length(sixth)>1){
mu6FrtC <- numeric(Nsim)
}
sigma1SqFrtC <- numeric(Nsim)
sigma2SqFrtC <- numeric(Nsim)
sigma3SqFrtC <- numeric(Nsim)
if(length(fourth)>1){
sigma4SqFrtC <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5SqFrtC <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6SqFrtC <- numeric(Nsim)
}
mu1[1]=mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2[1]=mean(secondComponent)
mu3[1]=mean(thirdComponent)
if(length(fourth)>1){
mu4[1]=mean(fourthComponent)
}
if(length(fifth)>1){
mu5[1]=mean(fifthComponent)
}
if(length(sixth)>1){
mu6[1]=mean(sixthComponent)
}
sigma1Sq[1]=var(firstComponent)
sigma2Sq[1]=var(secondComponent)
sigma3Sq[1]=var(thirdComponent)
if(length(fourth)>1){
sigma4Sq[1]=var(fourthComponent)
}
if(length(fifth)>1){
sigma5Sq[1]=var(fifthComponent)
}
if(length(sixth)>1){
sigma6Sq[1]=var(sixthComponent)
}
mu1SecC[1]=20*mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2SecC[1]=20*mean(secondComponent)
mu3SecC[1]=20*mean(thirdComponent)
if(length(fourth)>1){
mu4SecC[1]=20*mean(fourthComponent)
}
if(length(fifth)>1){
mu5SecC[1]=20*mean(fifthComponent)
}
if(length(sixth)>1){
mu6SecC[1]=20*mean(sixthComponent)
}
sigma1SqSecC[1]=20*var(firstComponent)
sigma2SqSecC[1]=20*var(secondComponent)
sigma3SqSecC[1]=20*var(thirdComponent)
if(length(fourth)>1){
sigma4SqSecC[1]=20*var(fourthComponent)
}
if(length(fifth)>1){
sigma5SqSecC[1]=20*var(fifthComponent)
}
if(length(sixth)>1){
sigma6SqSecC[1]=20*var(sixthComponent)
}
mu1ThdC[1]=1/20*mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2ThdC[1]=1/20*mean(secondComponent)
mu3ThdC[1]=1/20*mean(thirdComponent)
if(length(fourth)>1){
mu4ThdC[1]=1/20*mean(fourthComponent)
}
if(length(fifth)>1){
mu5ThdC[1]=1/20*mean(fifthComponent)
}
if(length(sixth)>1){
mu6ThdC[1]=1/20*mean(sixthComponent)
}
sigma1SqThdC[1]=1/20*var(firstComponent)
sigma2SqThdC[1]=1/20*var(secondComponent)
sigma3SqThdC[1]=1/20*var(thirdComponent)
if(length(fourth)>1){
sigma4SqThdC[1]=1/20*var(fourthComponent)
}
if(length(fifth)>1){
sigma5SqThdC[1]=1/20*var(fifthComponent)
}
if(length(sixth)>1){
sigma6SqThdC[1]=1/20*var(sixthComponent)
}
mu1FrtC[1]=10*mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2FrtC[1]=10*mean(secondComponent)
mu3FrtC[1]=10*mean(thirdComponent)
if(length(fourth)>1){
mu4FrtC[1]=10*mean(fourthComponent)
}
if(length(fifth)>1){
mu5FrtC[1]=10*mean(fifthComponent)
}
if(length(sixth)>1){
mu6FrtC[1]=10*mean(sixthComponent)
}
sigma1SqFrtC[1]=10*var(firstComponent)
sigma2SqFrtC[1]=10*var(secondComponent)
sigma3SqFrtC[1]=10*var(thirdComponent)
if(length(fourth)>1){
sigma4SqFrtC[1]=10*var(fourthComponent)
}
if(length(fifth)>1){
sigma5SqFrtC[1]=10*var(fifthComponent)
}
if(length(sixth)>1){
sigma6SqFrtC[1]=10*var(sixthComponent)
}
# Formula parts
N_1_S=length(firstComponent)
N_2_S=length(secondComponent)
N_3_S=length(thirdComponent)
if(length(fourth)>1){
N_4_S=length(fourthComponent)
}
if(length(fifth)>1){
N_5_S=length(fifthComponent)
}
if(length(sixth)>1){
N_6_S=length(sixthComponent)
}
vary1=var(firstComponent)
vary2=var(secondComponent)
vary3=var(thirdComponent)
if(length(fourth)>1){
vary4=var(fourthComponent)
}
if(length(fifth)>1){
vary5=var(fifthComponent)
}
if(length(sixth)>1){
vary6=var(sixthComponent)
}
bary1=mean(firstComponent)
bary2=mean(secondComponent)
bary3=mean(thirdComponent)
if(length(fourth)>1){
bary4=mean(fourthComponent)
}
if(length(fifth)>1){
bary5=mean(fifthComponent)
}
if(length(sixth)>1){
bary6=mean(sixthComponent)
}
# Set hyperparameters
if(hyperpars=="rafterys"){
if(length(first)>1 && length(second)>1 && length(third)>1){
smple=c(firstComponent,secondComponent,thirdComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent,sixthComponent)
}
b0=mean(smple) # Raftery's hyperparameters
B0=var(smple)/2.6*(max(smple)-min(smple))
#B0=sd(smple)*10
c0=1.28 # Raftery's hyperparameters
C0=0.36*var(smple) # Raftery's hyperparameters
#C0=10*sd(smple)
}
if(hyperpars=="custom"){
#b0=var(smple)*100*((mean(smple)/var(smple))*4) # Raftery's hyperparameters
#B0=var(smple)*100
#c0=1.28 # Raftery's hyperparameters
#C0=var(smple)*2.0
#b0=mean(smple) # Raftery's hyperparameters
#B0=var(smple)/0.8*(max(smple)-min(smple))
#c0=30*var(smple) # Raftery's hyperparameters
#C0=900*var(smple) # Raftery's hyperparameters
#b0=mean(smple) # Raftery's hyperparameters
#B0=100*var(smple)
#c0=50*var(smple) # Raftery's hyperparameters
#C0=900*var(smple) # Raftery's hyperparameters
# Funktionierte bisher ganz gut
if(length(first)>1 && length(second)>1 && length(third)>1){
smple=c(firstComponent,secondComponent,thirdComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent,sixthComponent)
}
b0=mean(smple) # Raftery's hyperparameters
B0=250*var(smple)
c0=1.28 # Raftery's hyperparameters
#C0=-0.5*var(smple)+(c0+0.5*(length(firstComp)+length(secondComp)))*5 # Raftery's hyperparameters
if(length(first)>1 && length(second)>1 && length(third)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)))*q
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)+length(fourthComponent)))*q
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)+length(fourthComponent)+length(fifthComponent)))*q
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)+length(fourthComponent)+length(fifthComponent)+length(sixthComponent)))*q
}
}
# Gibbs sampling via full conditionals
for(t in 2:Nsim){
# FIRST CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
c_1_S=c0+0.5*N_1_S
C_1_S=C0+0.5*sum((firstComponent-mu1[t-1])^2)
sigma1Sq[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
c_2_S=c0+0.5*N_2_S
C_2_S=C0+0.5*sum((secondComponent-mu2[t-1])^2)
sigma2Sq[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
c_3_S=c0+0.5*N_3_S
C_3_S=C0+0.5*sum((thirdComponent-mu3[t-1])^2)
sigma3Sq[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
c_4_S=c0+0.5*N_4_S
C_4_S=C0+0.5*sum((fourthComponent-mu4[t-1])^2)
sigma4Sq[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
c_5_S=c0+0.5*N_5_S
C_5_S=C0+0.5*sum((fifthComponent-mu5[t-1])^2)
sigma5Sq[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
c_6_S=c0+0.5*N_6_S
C_6_S=C0+0.5*sum((sixthComponent-mu6[t-1])^2)
sigma6Sq[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1Sq[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1Sq[t])*N_1_S*bary1+(1/B0)*b0)
mu1[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2Sq[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2Sq[t])*N_2_S*bary2+(1/B0)*b0)
mu2[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3Sq[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3Sq[t])*N_3_S*bary3+(1/B0)*b0)
mu3[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4Sq[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4Sq[t])*N_4_S*bary4+(1/B0)*b0)
mu4[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5Sq[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5Sq[t])*N_5_S*bary5+(1/B0)*b0)
mu5[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6Sq[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6Sq[t])*N_6_S*bary6+(1/B0)*b0)
mu6[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
###############################################################
# SECOND CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
C_1_S=C0+0.5*sum((firstComponent-mu1SecC[t-1])^2)
sigma1SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
C_2_S=C0+0.5*sum((secondComponent-mu2SecC[t-1])^2)
sigma2SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
C_3_S=C0+0.5*sum((thirdComponent-mu3SecC[t-1])^2)
sigma3SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
C_4_S=C0+0.5*sum((fourthComponent-mu4SecC[t-1])^2)
sigma4SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
C_5_S=C0+0.5*sum((fifthComponent-mu5SecC[t-1])^2)
sigma5SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
C_6_S=C0+0.5*sum((sixthComponent-mu6SecC[t-1])^2)
sigma6SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1SqSecC[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1SqSecC[t])*N_1_S*bary1+(1/B0)*b0)
mu1SecC[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2SqSecC[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2SqSecC[t])*N_2_S*bary2+(1/B0)*b0)
mu2SecC[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3SqSecC[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3SqSecC[t])*N_3_S*bary3+(1/B0)*b0)
mu3SecC[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4SqSecC[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4SqSecC[t])*N_4_S*bary4+(1/B0)*b0)
mu4SecC[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5SqSecC[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5SqSecC[t])*N_5_S*bary5+(1/B0)*b0)
mu5SecC[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6SqSecC[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6SqSecC[t])*N_6_S*bary6+(1/B0)*b0)
mu6SecC[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
###############################################################
# THIRD CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
C_1_S=C0+0.5*sum((firstComponent-mu1ThdC[t-1])^2)
sigma1SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
C_2_S=C0+0.5*sum((secondComponent-mu2ThdC[t-1])^2)
sigma2SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
C_3_S=C0+0.5*sum((thirdComponent-mu3ThdC[t-1])^2)
sigma3SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
C_4_S=C0+0.5*sum((fourthComponent-mu4ThdC[t-1])^2)
sigma4SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
C_5_S=C0+0.5*sum((fifthComponent-mu5ThdC[t-1])^2)
sigma5SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
C_6_S=C0+0.5*sum((sixthComponent-mu6ThdC[t-1])^2)
sigma6SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1SqThdC[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1SqThdC[t])*N_1_S*bary1+(1/B0)*b0)
mu1ThdC[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2SqThdC[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2SqThdC[t])*N_2_S*bary2+(1/B0)*b0)
mu2ThdC[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3SqThdC[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3SqThdC[t])*N_3_S*bary3+(1/B0)*b0)
mu3ThdC[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4SqThdC[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4SqThdC[t])*N_4_S*bary4+(1/B0)*b0)
mu4ThdC[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5SqThdC[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5SqThdC[t])*N_5_S*bary5+(1/B0)*b0)
mu5ThdC[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6SqThdC[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6SqThdC[t])*N_6_S*bary6+(1/B0)*b0)
mu6ThdC[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
###############################################################
# FOURTH CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
C_1_S=C0+0.5*sum((firstComponent-mu1FrtC[t-1])^2)
sigma1SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
C_2_S=C0+0.5*sum((secondComponent-mu2FrtC[t-1])^2)
sigma2SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
C_3_S=C0+0.5*sum((thirdComponent-mu3FrtC[t-1])^2)
sigma3SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
C_4_S=C0+0.5*sum((fourthComponent-mu4FrtC[t-1])^2)
sigma4SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
C_5_S=C0+0.5*sum((fifthComponent-mu5FrtC[t-1])^2)
sigma5SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
C_6_S=C0+0.5*sum((sixthComponent-mu6FrtC[t-1])^2)
sigma6SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1SqFrtC[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1SqFrtC[t])*N_1_S*bary1+(1/B0)*b0)
mu1FrtC[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2SqFrtC[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2SqFrtC[t])*N_2_S*bary2+(1/B0)*b0)
mu2FrtC[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3SqFrtC[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3SqFrtC[t])*N_3_S*bary3+(1/B0)*b0)
mu3FrtC[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4SqFrtC[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4SqFrtC[t])*N_4_S*bary4+(1/B0)*b0)
mu4FrtC[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5SqFrtC[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5SqFrtC[t])*N_5_S*bary5+(1/B0)*b0)
mu5FrtC[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6SqFrtC[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6SqFrtC[t])*N_6_S*bary6+(1/B0)*b0)
mu6FrtC[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
##############################################################
# All chains sample, reiterate
}
# Adapt chains for selected burnin
mu1=mu1[burnin:Nsim]
mu1SecC=mu1SecC[burnin:Nsim]
mu1ThdC=mu1ThdC[burnin:Nsim]
mu1FrtC=mu1FrtC[burnin:Nsim]
mu2=mu2[burnin:Nsim]
mu2SecC=mu2SecC[burnin:Nsim]
mu2ThdC=mu2ThdC[burnin:Nsim]
mu2FrtC=mu2FrtC[burnin:Nsim]
mu3=mu3[burnin:Nsim]
mu3SecC=mu3SecC[burnin:Nsim]
mu3ThdC=mu3ThdC[burnin:Nsim]
mu3FrtC=mu3FrtC[burnin:Nsim]
if(length(fourth)>1){
mu4=mu4[burnin:Nsim]
mu4SecC=mu4SecC[burnin:Nsim]
mu4ThdC=mu4ThdC[burnin:Nsim]
mu4FrtC=mu4FrtC[burnin:Nsim]
}
if(length(fifth)>1){
mu5=mu5[burnin:Nsim]
mu5SecC=mu5SecC[burnin:Nsim]
mu5ThdC=mu5ThdC[burnin:Nsim]
mu5FrtC=mu5FrtC[burnin:Nsim]
}
if(length(sixth)>1){
mu6=mu6[burnin:Nsim]
mu6SecC=mu6SecC[burnin:Nsim]
mu6ThdC=mu6ThdC[burnin:Nsim]
mu6FrtC=mu6FrtC[burnin:Nsim]
}
sigma1Sq=sigma1Sq[burnin:Nsim]
sigma1SqSecC=sigma1SqSecC[burnin:Nsim]
sigma1SqThdC=sigma1SqThdC[burnin:Nsim]
sigma1SqFrtC=sigma1SqFrtC[burnin:Nsim]
sigma2Sq=sigma2Sq[burnin:Nsim]
sigma2SqSecC=sigma2SqSecC[burnin:Nsim]
sigma2SqThdC=sigma2SqThdC[burnin:Nsim]
sigma2SqFrtC=sigma2SqFrtC[burnin:Nsim]
sigma3Sq=sigma3Sq[burnin:Nsim]
sigma3SqSecC=sigma3SqSecC[burnin:Nsim]
sigma3SqThdC=sigma3SqThdC[burnin:Nsim]
sigma3SqFrtC=sigma3SqFrtC[burnin:Nsim]
if(length(fourth)>1){
sigma4Sq=sigma4Sq[burnin:Nsim]
sigma4SqSecC=sigma4SqSecC[burnin:Nsim]
sigma4SqThdC=sigma4SqThdC[burnin:Nsim]
sigma4SqFrtC=sigma4SqFrtC[burnin:Nsim]
}
if(length(fifth)>1){
sigma5Sq=sigma5Sq[burnin:Nsim]
sigma5SqSecC=sigma5SqSecC[burnin:Nsim]
sigma5SqThdC=sigma5SqThdC[burnin:Nsim]
sigma5SqFrtC=sigma5SqFrtC[burnin:Nsim]
}
if(length(sixth)>1){
sigma6Sq=sigma6Sq[burnin:Nsim]
sigma6SqSecC=sigma6SqSecC[burnin:Nsim]
sigma6SqThdC=sigma6SqThdC[burnin:Nsim]
sigma6SqFrtC=sigma6SqFrtC[burnin:Nsim]
}
if(sd=="sd"){
sigma1Sq=sqrt(sigma1Sq)
sigma1SqSecC=sqrt(sigma1SqSecC)
sigma1SqThdC=sqrt(sigma1SqThdC)
sigma1SqFrtC=sqrt(sigma1SqFrtC)
sigma2Sq=sqrt(sigma2Sq)
sigma2SqSecC=sqrt(sigma2SqSecC)
sigma2SqThdC=sqrt(sigma2SqThdC)
sigma2SqFrtC=sqrt(sigma2SqFrtC)
sigma3Sq=sqrt(sigma3Sq)
sigma3SqSecC=sqrt(sigma3SqSecC)
sigma3SqThdC=sqrt(sigma3SqThdC)
sigma3SqFrtC=sqrt(sigma3SqFrtC)
if(length(fourth)>1){
sigma4Sq=sqrt(sigma4Sq)
sigma4SqSecC=sqrt(sigma4SqSecC)
sigma4SqThdC=sqrt(sigma4SqThdC)
sigma4SqFrtC=sqrt(sigma4SqFrtC)
}
if(length(fifth)>1){
sigma5Sq=sqrt(sigma5Sq)
sigma5SqSecC=sqrt(sigma5SqSecC)
sigma5SqThdC=sqrt(sigma5SqThdC)
sigma5SqFrtC=sqrt(sigma5SqFrtC)
}
if(length(sixth)>1){
sigma6Sq=sqrt(sigma6Sq)
sigma6SqSecC=sqrt(sigma6SqSecC)
sigma6SqThdC=sqrt(sigma6SqThdC)
sigma6SqFrtC=sqrt(sigma6SqFrtC)
}
}
diffOfMeansMu2MinusMu1=mu2-mu1
diffOfMeansMu2MinusMu1SecC=mu2SecC-mu1SecC
diffOfMeansMu2MinusMu1ThdC=mu2ThdC-mu1ThdC
diffOfMeansMu2MinusMu1FrtC=mu2FrtC-mu1FrtC
diffOfMeansMu3MinusMu1=mu3-mu1
diffOfMeansMu3MinusMu1SecC=mu3SecC-mu1SecC
diffOfMeansMu3MinusMu1ThdC=mu3ThdC-mu1ThdC
diffOfMeansMu3MinusMu1FrtC=mu3FrtC-mu1FrtC
diffOfMeansMu3MinusMu2=mu3-mu2
diffOfMeansMu3MinusMu2SecC=mu3SecC-mu2SecC
diffOfMeansMu3MinusMu2ThdC=mu3ThdC-mu2ThdC
diffOfMeansMu3MinusMu2FrtC=mu3FrtC-mu2FrtC
if(length(fourth)>1){
diffOfMeansMu4MinusMu1=mu4-mu1
diffOfMeansMu4MinusMu1SecC=mu4SecC-mu1SecC
diffOfMeansMu4MinusMu1ThdC=mu4ThdC-mu1ThdC
diffOfMeansMu4MinusMu1FrtC=mu4FrtC-mu1FrtC
diffOfMeansMu4MinusMu2=mu4-mu2
diffOfMeansMu4MinusMu2SecC=mu4SecC-mu2SecC
diffOfMeansMu4MinusMu2ThdC=mu4ThdC-mu2ThdC
diffOfMeansMu4MinusMu2FrtC=mu4FrtC-mu2FrtC
diffOfMeansMu4MinusMu3=mu4-mu3
diffOfMeansMu4MinusMu3SecC=mu4SecC-mu3SecC
diffOfMeansMu4MinusMu3ThdC=mu4ThdC-mu3ThdC
diffOfMeansMu4MinusMu3FrtC=mu4FrtC-mu3FrtC
}
if(length(fourth)>1 && length(fifth)>1){
diffOfMeansMu5MinusMu1=mu5-mu1
diffOfMeansMu5MinusMu1SecC=mu5SecC-mu1SecC
diffOfMeansMu5MinusMu1ThdC=mu5ThdC-mu1ThdC
diffOfMeansMu5MinusMu1FrtC=mu5FrtC-mu1FrtC
diffOfMeansMu5MinusMu2=mu5-mu2
diffOfMeansMu5MinusMu2SecC=mu5SecC-mu2SecC
diffOfMeansMu5MinusMu2ThdC=mu5ThdC-mu2ThdC
diffOfMeansMu5MinusMu2FrtC=mu5FrtC-mu2FrtC
diffOfMeansMu5MinusMu3=mu5-mu3
diffOfMeansMu5MinusMu3SecC=mu5SecC-mu3SecC
diffOfMeansMu5MinusMu3ThdC=mu5ThdC-mu3ThdC
diffOfMeansMu5MinusMu3FrtC=mu5FrtC-mu3FrtC
diffOfMeansMu5MinusMu4=mu5-mu4
diffOfMeansMu5MinusMu4SecC=mu5SecC-mu4SecC
diffOfMeansMu5MinusMu4ThdC=mu5ThdC-mu4ThdC
diffOfMeansMu5MinusMu4FrtC=mu5FrtC-mu4FrtC
}
if(length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
diffOfMeansMu6MinusMu1=mu6-mu1
diffOfMeansMu6MinusMu1SecC=mu6SecC-mu1SecC
diffOfMeansMu6MinusMu1ThdC=mu6ThdC-mu1ThdC
diffOfMeansMu6MinusMu1FrtC=mu6FrtC-mu1FrtC
diffOfMeansMu6MinusMu2=mu6-mu2
diffOfMeansMu6MinusMu2SecC=mu6SecC-mu2SecC
diffOfMeansMu6MinusMu2ThdC=mu6ThdC-mu2ThdC
diffOfMeansMu6MinusMu2FrtC=mu6FrtC-mu2FrtC
diffOfMeansMu6MinusMu3=mu6-mu3
diffOfMeansMu6MinusMu3SecC=mu6SecC-mu3SecC
diffOfMeansMu6MinusMu3ThdC=mu6ThdC-mu3ThdC
diffOfMeansMu6MinusMu3FrtC=mu6FrtC-mu3FrtC
diffOfMeansMu6MinusMu4=mu6-mu4
diffOfMeansMu6MinusMu4SecC=mu6SecC-mu4SecC
diffOfMeansMu6MinusMu4ThdC=mu6ThdC-mu4ThdC
diffOfMeansMu6MinusMu4FrtC=mu6FrtC-mu4FrtC
diffOfMeansMu6MinusMu5=mu6-mu5
diffOfMeansMu6MinusMu5SecC=mu6SecC-mu5SecC
diffOfMeansMu6MinusMu5ThdC=mu6ThdC-mu5ThdC
diffOfMeansMu6MinusMu5FrtC=mu6FrtC-mu5FrtC
}
diffOfVariancesS2MinusS1=sigma2Sq-sigma1Sq
diffOfVariancesS2MinusS1SecC=sigma2SqSecC-sigma1SqSecC
diffOfVariancesS2MinusS1ThdC=sigma2SqThdC-sigma1SqThdC
diffOfVariancesS2MinusS1FrtC=sigma2SqFrtC-sigma1SqFrtC
diffOfVariancesS3MinusS1=sigma3Sq-sigma1Sq
diffOfVariancesS3MinusS1SecC=sigma3SqSecC-sigma1SqSecC
diffOfVariancesS3MinusS1ThdC=sigma3SqThdC-sigma1SqThdC
diffOfVariancesS3MinusS1FrtC=sigma3SqFrtC-sigma1SqFrtC
diffOfVariancesS3MinusS2=sigma3Sq-sigma2Sq
diffOfVariancesS3MinusS2SecC=sigma3SqSecC-sigma2SqSecC
diffOfVariancesS3MinusS2ThdC=sigma3SqThdC-sigma2SqThdC
diffOfVariancesS3MinusS2FrtC=sigma3SqFrtC-sigma2SqFrtC
if(length(fourth)>1){
diffOfVariancesS4MinusS1=sigma4Sq-sigma1Sq
diffOfVariancesS4MinusS1SecC=sigma4SqSecC-sigma1SqSecC
diffOfVariancesS4MinusS1ThdC=sigma4SqThdC-sigma1SqThdC
diffOfVariancesS4MinusS1FrtC=sigma4SqFrtC-sigma1SqFrtC
diffOfVariancesS4MinusS2=sigma4Sq-sigma2Sq
diffOfVariancesS4MinusS2SecC=sigma4SqSecC-sigma2SqSecC
diffOfVariancesS4MinusS2ThdC=sigma4SqThdC-sigma2SqThdC
diffOfVariancesS4MinusS2FrtC=sigma4SqFrtC-sigma2SqFrtC
diffOfVariancesS4MinusS3=sigma4Sq-sigma3Sq
diffOfVariancesS4MinusS3SecC=sigma4SqSecC-sigma3SqSecC
diffOfVariancesS4MinusS3ThdC=sigma4SqThdC-sigma3SqThdC
diffOfVariancesS4MinusS3FrtC=sigma4SqFrtC-sigma3SqFrtC
}
if(length(fourth)>1 && length(fifth)>1){
diffOfVariancesS5MinusS1=sigma5Sq-sigma1Sq
diffOfVariancesS5MinusS1SecC=sigma5SqSecC-sigma1SqSecC
diffOfVariancesS5MinusS1ThdC=sigma5SqThdC-sigma1SqThdC
diffOfVariancesS5MinusS1FrtC=sigma5SqFrtC-sigma1SqFrtC
diffOfVariancesS5MinusS2=sigma5Sq-sigma2Sq
diffOfVariancesS5MinusS2SecC=sigma5SqSecC-sigma2SqSecC
diffOfVariancesS5MinusS2ThdC=sigma5SqThdC-sigma2SqThdC
diffOfVariancesS5MinusS2FrtC=sigma5SqFrtC-sigma2SqFrtC
diffOfVariancesS5MinusS3=sigma5Sq-sigma3Sq
diffOfVariancesS5MinusS3SecC=sigma5SqSecC-sigma3SqSecC
diffOfVariancesS5MinusS3ThdC=sigma5SqThdC-sigma3SqThdC
diffOfVariancesS5MinusS3FrtC=sigma5SqFrtC-sigma3SqFrtC
diffOfVariancesS5MinusS4=sigma5Sq-sigma4Sq
diffOfVariancesS5MinusS4SecC=sigma5SqSecC-sigma4SqSecC
diffOfVariancesS5MinusS4ThdC=sigma5SqThdC-sigma4SqThdC
diffOfVariancesS5MinusS4FrtC=sigma5SqFrtC-sigma4SqFrtC
}
if(length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
diffOfVariancesS6MinusS1=sigma6Sq-sigma1Sq
diffOfVariancesS6MinusS1SecC=sigma6SqSecC-sigma1SqSecC
diffOfVariancesS6MinusS1ThdC=sigma6SqThdC-sigma1SqThdC
diffOfVariancesS6MinusS1FrtC=sigma6SqFrtC-sigma1SqFrtC
diffOfVariancesS6MinusS2=sigma6Sq-sigma2Sq
diffOfVariancesS6MinusS2SecC=sigma6SqSecC-sigma2SqSecC
diffOfVariancesS6MinusS2ThdC=sigma6SqThdC-sigma2SqThdC
diffOfVariancesS6MinusS2FrtC=sigma6SqFrtC-sigma2SqFrtC
diffOfVariancesS6MinusS3=sigma6Sq-sigma3Sq
diffOfVariancesS6MinusS3SecC=sigma6SqSecC-sigma3SqSecC
diffOfVariancesS6MinusS3ThdC=sigma6SqThdC-sigma3SqThdC
diffOfVariancesS6MinusS3FrtC=sigma6SqFrtC-sigma3SqFrtC
diffOfVariancesS6MinusS4=sigma6Sq-sigma4Sq
diffOfVariancesS6MinusS4SecC=sigma6SqSecC-sigma4SqSecC
diffOfVariancesS6MinusS4ThdC=sigma6SqThdC-sigma4SqThdC
diffOfVariancesS6MinusS4FrtC=sigma6SqFrtC-sigma4SqFrtC
diffOfVariancesS6MinusS5=sigma6Sq-sigma5Sq
diffOfVariancesS6MinusS5SecC=sigma6SqSecC-sigma5SqSecC
diffOfVariancesS6MinusS5ThdC=sigma6SqThdC-sigma5SqThdC
diffOfVariancesS6MinusS5FrtC=sigma6SqFrtC-sigma5SqFrtC
}
# Effect sizes
if(sd=="var"){
effectSize21=(mu1-mu2)/(sqrt(((N_1_S-1)*sigma1Sq+(N_2_S-1)*sigma2Sq)/(N_1_S+N_2_S-2)))
effectSize21SecC=(mu1SecC-mu2SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_2_S-1)*sigma2SqSecC)/(N_1_S+N_2_S-2)))
effectSize21ThdC=(mu1ThdC-mu2ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_2_S-1)*sigma2SqThdC)/(N_1_S+N_2_S-2)))
effectSize21FrtC=(mu1FrtC-mu2FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_2_S-1)*sigma2SqFrtC)/(N_1_S+N_2_S-2)))
effectSize13=(mu1-mu3)/(sqrt(((N_1_S-1)*sigma1Sq+(N_3_S-1)*sigma3Sq)/(N_1_S+N_3_S-2)))
effectSize13SecC=(mu1SecC-mu3SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_1_S+N_3_S-2)))
effectSize13ThdC=(mu1ThdC-mu3ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_1_S+N_3_S-2)))
effectSize13FrtC=(mu1FrtC-mu3FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_1_S+N_3_S-2)))
effectSize23=(mu2-mu3)/(sqrt(((N_2_S-1)*sigma2Sq+(N_3_S-1)*sigma3Sq)/(N_2_S+N_3_S-2)))
effectSize23SecC=(mu2SecC-mu3SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_2_S+N_3_S-2)))
effectSize23ThdC=(mu2ThdC-mu3ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_2_S+N_3_S-2)))
effectSize23FrtC=(mu2FrtC-mu3FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_2_S+N_3_S-2)))
if(length(fourth>1)){
effectSize14=(mu1-mu4)/(sqrt(((N_1_S-1)*sigma1Sq+(N_4_S-1)*sigma4Sq)/(N_1_S+N_4_S-2)))
effectSize14SecC=(mu1SecC-mu4SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_1_S+N_4_S-2)))
effectSize14ThdC=(mu1ThdC-mu4ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_1_S+N_4_S-2)))
effectSize14FrtC=(mu1FrtC-mu4FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_1_S+N_4_S-2)))
effectSize24=(mu2-mu4)/(sqrt(((N_2_S-1)*sigma2Sq+(N_4_S-1)*sigma4Sq)/(N_2_S+N_4_S-2)))
effectSize24SecC=(mu2SecC-mu4SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_2_S+N_4_S-2)))
effectSize24ThdC=(mu2ThdC-mu4ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_2_S+N_4_S-2)))
effectSize24FrtC=(mu2FrtC-mu4FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_2_S+N_4_S-2)))
effectSize34=(mu3-mu4)/(sqrt(((N_3_S-1)*sigma3Sq+(N_4_S-1)*sigma4Sq)/(N_3_S+N_4_S-2)))
effectSize34SecC=(mu3SecC-mu4SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_3_S+N_4_S-2)))
effectSize34ThdC=(mu3ThdC-mu4ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_3_S+N_4_S-2)))
effectSize34FrtC=(mu3FrtC-mu4FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_3_S+N_4_S-2)))
}
if(length(fourth>1) && length(fifth)>1){
effectSize15=(mu1-mu5)/(sqrt(((N_1_S-1)*sigma1Sq+(N_5_S-1)*sigma5Sq)/(N_1_S+N_5_S-2)))
effectSize15SecC=(mu1SecC-mu5SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_1_S+N_5_S-2)))
effectSize15ThdC=(mu1ThdC-mu5ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_1_S+N_5_S-2)))
effectSize15FrtC=(mu1FrtC-mu5FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_1_S+N_5_S-2)))
effectSize25=(mu2-mu5)/(sqrt(((N_2_S-1)*sigma2Sq+(N_5_S-1)*sigma5Sq)/(N_2_S+N_5_S-2)))
effectSize25SecC=(mu2SecC-mu5SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_2_S+N_5_S-2)))
effectSize25ThdC=(mu2ThdC-mu5ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_2_S+N_5_S-2)))
effectSize25FrtC=(mu2FrtC-mu5FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_2_S+N_5_S-2)))
effectSize35=(mu3-mu5)/(sqrt(((N_3_S-1)*sigma3Sq+(N_5_S-1)*sigma5Sq)/(N_3_S+N_5_S-2)))
effectSize35SecC=(mu3SecC-mu5SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_3_S+N_5_S-2)))
effectSize35ThdC=(mu3ThdC-mu5ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_3_S+N_5_S-2)))
effectSize35FrtC=(mu3FrtC-mu5FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_3_S+N_5_S-2)))
effectSize45=(mu4-mu5)/(sqrt(((N_4_S-1)*sigma4Sq+(N_5_S-1)*sigma5Sq)/(N_4_S+N_5_S-2)))
effectSize45SecC=(mu4SecC-mu5SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_4_S+N_5_S-2)))
effectSize45ThdC=(mu4ThdC-mu5ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_4_S+N_5_S-2)))
effectSize45FrtC=(mu4FrtC-mu5FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_4_S+N_5_S-2)))
}
if(length(fourth>1)&& length(fifth)>1 && length(sixth)>1){
effectSize16=(mu1-mu6)/(sqrt(((N_1_S-1)*sigma1Sq+(N_6_S-1)*sigma6Sq)/(N_1_S+N_6_S-2)))
effectSize16SecC=(mu1SecC-mu6SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_1_S+N_6_S-2)))
effectSize16ThdC=(mu1ThdC-mu6ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_1_S+N_6_S-2)))
effectSize16FrtC=(mu1FrtC-mu6FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_1_S+N_6_S-2)))
effectSize26=(mu2-mu6)/(sqrt(((N_2_S-1)*sigma2Sq+(N_6_S-1)*sigma6Sq)/(N_2_S+N_6_S-2)))
effectSize26SecC=(mu2SecC-mu6SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_2_S+N_6_S-2)))
effectSize26ThdC=(mu2ThdC-mu6ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_2_S+N_6_S-2)))
effectSize26FrtC=(mu2FrtC-mu6FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_2_S+N_6_S-2)))
effectSize36=(mu3-mu6)/(sqrt(((N_3_S-1)*sigma3Sq+(N_6_S-1)*sigma6Sq)/(N_3_S+N_6_S-2)))
effectSize36SecC=(mu3SecC-mu6SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_3_S+N_6_S-2)))
effectSize36ThdC=(mu3ThdC-mu6ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_3_S+N_6_S-2)))
effectSize36FrtC=(mu3FrtC-mu6FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_3_S+N_6_S-2)))
effectSize46=(mu4-mu6)/(sqrt(((N_4_S-1)*sigma4Sq+(N_6_S-1)*sigma6Sq)/(N_4_S+N_6_S-2)))
effectSize46SecC=(mu4SecC-mu6SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_4_S+N_6_S-2)))
effectSize46ThdC=(mu4ThdC-mu6ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_4_S+N_6_S-2)))
effectSize46FrtC=(mu4FrtC-mu6FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_4_S+N_6_S-2)))
effectSize56=(mu5-mu6)/(sqrt(((N_5_S-1)*sigma5Sq+(N_6_S-1)*sigma6Sq)/(N_5_S+N_6_S-2)))
effectSize56SecC=(mu5SecC-mu6SecC)/(sqrt(((N_5_S-1)*sigma5SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_5_S+N_6_S-2)))
effectSize56ThdC=(mu5ThdC-mu6ThdC)/(sqrt(((N_5_S-1)*sigma5SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_5_S+N_6_S-2)))
effectSize56FrtC=(mu5FrtC-mu6FrtC)/(sqrt(((N_5_S-1)*sigma5SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_5_S+N_6_S-2)))
}
}
if(sd=="sd"){
effectSize12=(mu1-mu2)/(sqrt(((N_1_S-1)*sigma1Sq+(N_2_S-1)*sigma2Sq)/(N_1_S+N_2_S-2)))
effectSize12SecC=(mu1SecC-mu2SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_2_S-1)*sigma2SqSecC)/(N_1_S+N_2_S-2)))
effectSize12ThdC=(mu1ThdC-mu2ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_2_S-1)*sigma2SqThdC)/(N_1_S+N_2_S-2)))
effectSize12FrtC=(mu1FrtC-mu2FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_2_S-1)*sigma2SqFrtC)/(N_1_S+N_2_S-2)))
effectSize13=(mu1-mu3)/(sqrt(((N_1_S-1)*sigma1Sq+(N_3_S-1)*sigma3Sq)/(N_1_S+N_3_S-2)))
effectSize13SecC=(mu1SecC-mu3SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_1_S+N_3_S-2)))
effectSize13ThdC=(mu1ThdC-mu3ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_1_S+N_3_S-2)))
effectSize13FrtC=(mu1FrtC-mu3FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_1_S+N_3_S-2)))
effectSize23=(mu2-mu3)/(sqrt(((N_2_S-1)*sigma2Sq+(N_3_S-1)*sigma3Sq)/(N_2_S+N_3_S-2)))
effectSize23SecC=(mu2SecC-mu3SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_2_S+N_3_S-2)))
effectSize23ThdC=(mu2ThdC-mu3ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_2_S+N_3_S-2)))
effectSize23FrtC=(mu2FrtC-mu3FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_2_S+N_3_S-2)))
if(length(fourth>1)){
effectSize14=(mu1-mu4)/(sqrt(((N_1_S-1)*sigma1Sq+(N_4_S-1)*sigma4Sq)/(N_1_S+N_4_S-2)))
effectSize14SecC=(mu1SecC-mu4SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_1_S+N_4_S-2)))
effectSize14ThdC=(mu1ThdC-mu4ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_1_S+N_4_S-2)))
effectSize14FrtC=(mu1FrtC-mu4FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_1_S+N_4_S-2)))
effectSize24=(mu2-mu4)/(sqrt(((N_2_S-1)*sigma2Sq+(N_4_S-1)*sigma4Sq)/(N_2_S+N_4_S-2)))
effectSize24SecC=(mu2SecC-mu4SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_2_S+N_4_S-2)))
effectSize24ThdC=(mu2ThdC-mu4ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_2_S+N_4_S-2)))
effectSize24FrtC=(mu2FrtC-mu4FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_2_S+N_4_S-2)))
effectSize34=(mu3-mu4)/(sqrt(((N_3_S-1)*sigma3Sq+(N_4_S-1)*sigma4Sq)/(N_3_S+N_4_S-2)))
effectSize34SecC=(mu3SecC-mu4SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_3_S+N_4_S-2)))
effectSize34ThdC=(mu3ThdC-mu4ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_3_S+N_4_S-2)))
effectSize34FrtC=(mu3FrtC-mu4FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_3_S+N_4_S-2)))
}
if(length(fourth>1) && length(fifth)>1){
effectSize15=(mu1-mu5)/(sqrt(((N_1_S-1)*sigma1Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_1_S+N_5_S-2)))
effectSize15SecC=(mu1SecC-mu5SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_1_S+N_5_S-2)))
effectSize15ThdC=(mu1ThdC-mu5ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_1_S+N_5_S-2)))
effectSize15FrtC=(mu1FrtC-mu5FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_1_S+N_5_S-2)))
effectSize25=(mu2-mu5)/(sqrt(((N_2_S-1)*sigma2Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_2_S+N_5_S-2)))
effectSize25SecC=(mu2SecC-mu5SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_2_S+N_5_S-2)))
effectSize25ThdC=(mu2ThdC-mu5ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_2_S+N_5_S-2)))
effectSize25FrtC=(mu2FrtC-mu5FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_2_S+N_5_S-2)))
effectSize35=(mu3-mu5)/(sqrt(((N_3_S-1)*sigma3Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_3_S+N_5_S-2)))
effectSize35SecC=(mu3SecC-mu5SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_3_S+N_5_S-2)))
effectSize35ThdC=(mu3ThdC-mu5ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_3_S+N_5_S-2)))
effectSize35FrtC=(mu3FrtC-mu5FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_3_S+N_5_S-2)))
effectSize45=(mu4-mu5)/(sqrt(((N_4_S-1)*sigma4Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_4_S+N_5_S-2)))
effectSize45SecC=(mu4SecC-mu5SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_4_S+N_5_S-2)))
effectSize45ThdC=(mu4ThdC-mu5ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_4_S+N_5_S-2)))
effectSize45FrtC=(mu4FrtC-mu5FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_4_S+N_5_S-2)))
}
if(length(fourth>1)&& length(fifth)>1 && length(sixth)>1){
effectSize16=(mu1-mu6)/(sqrt(((N_1_S-1)*sigma1Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_1_S+N_6_S-2)))
effectSize16SecC=(mu1SecC-mu6SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_1_S+N_6_S-2)))
effectSize16ThdC=(mu1ThdC-mu6ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_1_S+N_6_S-2)))
effectSize16FrtC=(mu1FrtC-mu6FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_1_S+N_6_S-2)))
effectSize26=(mu2-mu6)/(sqrt(((N_2_S-1)*sigma2Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_2_S+N_6_S-2)))
effectSize26SecC=(mu2SecC-mu6SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_2_S+N_6_S-2)))
effectSize26ThdC=(mu2ThdC-mu6ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_2_S+N_6_S-2)))
effectSize26FrtC=(mu2FrtC-mu6FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_2_S+N_6_S-2)))
effectSize36=(mu3-mu6)/(sqrt(((N_3_S-1)*sigma3Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_3_S+N_6_S-2)))
effectSize36SecC=(mu3SecC-mu6SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_3_S+N_6_S-2)))
effectSize36ThdC=(mu3ThdC-mu6ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_3_S+N_6_S-2)))
effectSize36FrtC=(mu3FrtC-mu6FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_3_S+N_6_S-2)))
effectSize46=(mu4-mu6)/(sqrt(((N_4_S-1)*sigma4Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_4_S+N_6_S-2)))
effectSize46SecC=(mu4SecC-mu6SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_4_S+N_6_S-2)))
effectSize46ThdC=(mu4ThdC-mu6ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_4_S+N_6_S-2)))
effectSize46FrtC=(mu4FrtC-mu6FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_4_S+N_6_S-2)))
effectSize56=(mu5-mu6)/(sqrt(((N_5_S-1)*sigma5Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_5_S+N_6_S-2)))
effectSize56SecC=(mu5SecC-mu6SecC)/(sqrt(((N_5_S-1)*sigma5SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_5_S+N_6_S-2)))
effectSize56ThdC=(mu5ThdC-mu6ThdC)/(sqrt(((N_5_S-1)*sigma5SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_5_S+N_6_S-2)))
effectSize56FrtC=(mu5FrtC-mu6FrtC)/(sqrt(((N_5_S-1)*sigma5SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_5_S+N_6_S-2)))
}
}
##################################### CONSOLE OUTPUT ############################################
message("Bayesian ANOVA output:")
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1) {
message("Details: Gaussian-mixture model with six components")
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1) {
message("Details: Gaussian-mixture model with five components")
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1) {
message("Details: Gaussian-mixture model with four components")
} else if(length(first)>1 && length(second)>1 && length(third)>1) {
message("Details: Gaussian-mixture model with three components")
}
# Quantiles
mu1Quants=quantile(mu1,probs=c((1-ci)/2,ci+(1-ci)/2))
mu2Quants=quantile(mu2,probs=c((1-ci)/2,ci+(1-ci)/2))
mu3Quants=quantile(mu3,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
mu4Quants=quantile(mu4,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
mu5Quants=quantile(mu5,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
mu6Quants=quantile(mu6,probs=c((1-ci)/2,ci+(1-ci)/2))
}
sigma1Quants=quantile(sigma1Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
sigma2Quants=quantile(sigma2Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
sigma3Quants=quantile(sigma3Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
sigma4Quants=quantile(sigma4Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
sigma5Quants=quantile(sigma5Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
sigma6Quants=quantile(sigma6Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
}
diffOfMeansMu2MinusMu1Quants=quantile(diffOfMeansMu2MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu3MinusMu1Quants=quantile(diffOfMeansMu3MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu3MinusMu2Quants=quantile(diffOfMeansMu3MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
diffOfMeansMu4MinusMu1Quants=quantile(diffOfMeansMu4MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu4MinusMu2Quants=quantile(diffOfMeansMu4MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu4MinusMu3Quants=quantile(diffOfMeansMu4MinusMu3,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
diffOfMeansMu5MinusMu1Quants=quantile(diffOfMeansMu5MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu5MinusMu2Quants=quantile(diffOfMeansMu5MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu5MinusMu3Quants=quantile(diffOfMeansMu5MinusMu3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu5MinusMu4Quants=quantile(diffOfMeansMu5MinusMu4,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
diffOfMeansMu6MinusMu1Quants=quantile(diffOfMeansMu6MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu2Quants=quantile(diffOfMeansMu6MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu3Quants=quantile(diffOfMeansMu6MinusMu3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu4Quants=quantile(diffOfMeansMu6MinusMu4,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu5Quants=quantile(diffOfMeansMu6MinusMu5,probs=c((1-ci)/2,ci+(1-ci)/2))
}
diffOfVariancesS2MinusS1Quants=quantile(diffOfVariancesS2MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS3MinusS1Quants=quantile(diffOfVariancesS3MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS3MinusS2Quants=quantile(diffOfVariancesS3MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
diffOfVariancesS4MinusS1Quants=quantile(diffOfVariancesS4MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS4MinusS2Quants=quantile(diffOfVariancesS4MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS4MinusS3Quants=quantile(diffOfVariancesS4MinusS3,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
diffOfVariancesS5MinusS1Quants=quantile(diffOfVariancesS5MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS5MinusS2Quants=quantile(diffOfVariancesS5MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS5MinusS3Quants=quantile(diffOfVariancesS5MinusS3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS5MinusS4Quants=quantile(diffOfVariancesS5MinusS4,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
diffOfVariancesS6MinusS1Quants=quantile(diffOfVariancesS6MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS2Quants=quantile(diffOfVariancesS6MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS3Quants=quantile(diffOfVariancesS6MinusS3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS4Quants=quantile(diffOfVariancesS6MinusS4,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS5Quants=quantile(diffOfVariancesS6MinusS5,probs=c((1-ci)/2,ci+(1-ci)/2))
}
effectSize12Quants=quantile(effectSize12,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize13Quants=quantile(effectSize13,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize23Quants=quantile(effectSize23,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
effectSize14Quants=quantile(effectSize14,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize24Quants=quantile(effectSize24,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize34Quants=quantile(effectSize34,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
effectSize15Quants=quantile(effectSize15,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize25Quants=quantile(effectSize25,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize35Quants=quantile(effectSize35,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize45Quants=quantile(effectSize45,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
effectSize16Quants=quantile(effectSize16,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize26Quants=quantile(effectSize26,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize36Quants=quantile(effectSize36,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize46Quants=quantile(effectSize46,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize56Quants=quantile(effectSize56,probs=c((1-ci)/2,ci+(1-ci)/2))
}
# Console Output
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1) {
Parameter<-c("mu1",
"mu2",
"mu3",
"mu4",
"mu5",
"mu6",
"sigma1",
"sigma2",
"sigma3",
"sigma4",
"sigma5",
"sigma6",
"mu2-mu1",
"mu3-mu1",
"mu4-mu1",
"mu5-mu1",
"mu6-mu1",
"mu3-mu2",
"mu4-mu2",
"mu5-mu2",
"mu6-mu2",
"mu4-mu3",
"mu5-mu3",
"mu6-mu3",
"mu5-mu4",
"mu6-mu4",
"mu6-mu5",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma4-sigma1",
"sigma5-sigma1",
"sigma6-sigma1",
"sigma3-sigma2",
"sigma4-sigma2",
"sigma5-sigma2",
"sigma6-sigma2",
"sigma4-sigma3",
"sigma5-sigma3",
"sigma6-sigma3",
"sigma5-sigma4",
"sigma6-sigma4",
"sigma6-sigma5",
"delta12",
"delta13",
"delta14",
"delta15",
"delta16",
"delta23",
"delta24",
"delta25",
"delta26",
"delta34",
"delta35",
"delta36",
"delta45",
"delta46",
"delta56")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(mu4Quants[1]),
as.numeric(mu5Quants[1]),
as.numeric(mu6Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(sigma4Quants[1]),
as.numeric(sigma5Quants[1]),
as.numeric(sigma6Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu4MinusMu1Quants[1]),
as.numeric(diffOfMeansMu5MinusMu1Quants[1]),
as.numeric(diffOfMeansMu6MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu2Quants[1]),
as.numeric(diffOfMeansMu5MinusMu2Quants[1]),
as.numeric(diffOfMeansMu6MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu3Quants[1]),
as.numeric(diffOfMeansMu6MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu4Quants[1]),
as.numeric(diffOfMeansMu6MinusMu4Quants[1]),
as.numeric(diffOfMeansMu6MinusMu5Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS4MinusS1Quants[1]),
as.numeric(diffOfVariancesS5MinusS1Quants[1]),
as.numeric(diffOfVariancesS6MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS2Quants[1]),
as.numeric(diffOfVariancesS5MinusS2Quants[1]),
as.numeric(diffOfVariancesS6MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS3Quants[1]),
as.numeric(diffOfVariancesS6MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS4Quants[1]),
as.numeric(diffOfVariancesS6MinusS4Quants[1]),
as.numeric(diffOfVariancesS6MinusS5Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize14Quants[1]),
as.numeric(effectSize15Quants[1]),
as.numeric(effectSize16Quants[1]),
as.numeric(effectSize23Quants[1]),
as.numeric(effectSize24Quants[1]),
as.numeric(effectSize25Quants[1]),
as.numeric(effectSize26Quants[1]),
as.numeric(effectSize34Quants[1]),
as.numeric(effectSize35Quants[1]),
as.numeric(effectSize36Quants[1]),
as.numeric(effectSize45Quants[1]),
as.numeric(effectSize46Quants[1]),
as.numeric(effectSize56Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(mu4),
mean(mu5),
mean(mu6),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(sigma4Sq),
mean(sigma5Sq),
mean(sigma6Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu4MinusMu1),
mean(diffOfMeansMu5MinusMu1),
mean(diffOfMeansMu6MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfMeansMu4MinusMu2),
mean(diffOfMeansMu5MinusMu2),
mean(diffOfMeansMu6MinusMu2),
mean(diffOfMeansMu4MinusMu3),
mean(diffOfMeansMu5MinusMu3),
mean(diffOfMeansMu6MinusMu3),
mean(diffOfMeansMu5MinusMu4),
mean(diffOfMeansMu6MinusMu4),
mean(diffOfMeansMu6MinusMu5),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS4MinusS1),
mean(diffOfVariancesS5MinusS1),
mean(diffOfVariancesS6MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(diffOfVariancesS4MinusS2),
mean(diffOfVariancesS5MinusS2),
mean(diffOfVariancesS6MinusS2),
mean(diffOfVariancesS4MinusS3),
mean(diffOfVariancesS5MinusS3),
mean(diffOfVariancesS6MinusS3),
mean(diffOfVariancesS5MinusS4),
mean(diffOfVariancesS6MinusS4),
mean(diffOfVariancesS6MinusS5),
mean(effectSize12),
mean(effectSize13),
mean(effectSize14),
mean(effectSize15),
mean(effectSize16),
mean(effectSize23),
mean(effectSize24),
mean(effectSize25),
mean(effectSize26),
mean(effectSize34),
mean(effectSize35),
mean(effectSize36),
mean(effectSize45),
mean(effectSize46),
mean(effectSize56)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(mu4Quants[2]),
as.numeric(mu5Quants[2]),
as.numeric(mu6Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(sigma4Quants[2]),
as.numeric(sigma5Quants[2]),
as.numeric(sigma6Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu4MinusMu1Quants[2]),
as.numeric(diffOfMeansMu5MinusMu1Quants[2]),
as.numeric(diffOfMeansMu6MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu2Quants[2]),
as.numeric(diffOfMeansMu5MinusMu2Quants[2]),
as.numeric(diffOfMeansMu6MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu3Quants[2]),
as.numeric(diffOfMeansMu6MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu4Quants[2]),
as.numeric(diffOfMeansMu6MinusMu4Quants[2]),
as.numeric(diffOfMeansMu6MinusMu5Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS4MinusS1Quants[2]),
as.numeric(diffOfVariancesS5MinusS1Quants[2]),
as.numeric(diffOfVariancesS6MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS2Quants[2]),
as.numeric(diffOfVariancesS5MinusS2Quants[2]),
as.numeric(diffOfVariancesS6MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS3Quants[2]),
as.numeric(diffOfVariancesS6MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS4Quants[2]),
as.numeric(diffOfVariancesS6MinusS4Quants[2]),
as.numeric(diffOfVariancesS6MinusS5Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize14Quants[2]),
as.numeric(effectSize15Quants[2]),
as.numeric(effectSize16Quants[2]),
as.numeric(effectSize23Quants[2]),
as.numeric(effectSize24Quants[2]),
as.numeric(effectSize25Quants[2]),
as.numeric(effectSize26Quants[2]),
as.numeric(effectSize34Quants[2]),
as.numeric(effectSize35Quants[2]),
as.numeric(effectSize36Quants[2]),
as.numeric(effectSize45Quants[2]),
as.numeric(effectSize46Quants[2]),
as.numeric(effectSize56Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(mu4),
sd(mu5),
sd(mu6),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(sigma4Sq),
sd(sigma5Sq),
sd(sigma6Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu4MinusMu1),
sd(diffOfMeansMu5MinusMu1),
sd(diffOfMeansMu6MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfMeansMu4MinusMu2),
sd(diffOfMeansMu5MinusMu2),
sd(diffOfMeansMu6MinusMu2),
sd(diffOfMeansMu4MinusMu3),
sd(diffOfMeansMu5MinusMu3),
sd(diffOfMeansMu6MinusMu3),
sd(diffOfMeansMu5MinusMu4),
sd(diffOfMeansMu6MinusMu4),
sd(diffOfMeansMu6MinusMu5),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS4MinusS1),
sd(diffOfVariancesS5MinusS1),
sd(diffOfVariancesS6MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(diffOfVariancesS4MinusS2),
sd(diffOfVariancesS5MinusS2),
sd(diffOfVariancesS6MinusS2),
sd(diffOfVariancesS4MinusS3),
sd(diffOfVariancesS5MinusS3),
sd(diffOfVariancesS6MinusS3),
sd(diffOfVariancesS5MinusS4),
sd(diffOfVariancesS6MinusS4),
sd(diffOfVariancesS6MinusS5),
sd(effectSize12),
sd(effectSize13),
sd(effectSize14),
sd(effectSize15),
sd(effectSize16),
sd(effectSize23),
sd(effectSize24),
sd(effectSize25),
sd(effectSize26),
sd(effectSize34),
sd(effectSize35),
sd(effectSize36),
sd(effectSize45),
sd(effectSize46),
sd(effectSize56)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
# posterior draw dataframe
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
mu4, mu4SecC, mu4ThdC, mu4FrtC,
mu5, mu5SecC, mu5ThdC, mu5FrtC,
mu6, mu6SecC, mu6ThdC, mu6FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
sigma4Sq, sigma4SqSecC, sigma4SqThdC, sigma4SqFrtC,
sigma5Sq, sigma5SqSecC, sigma5SqThdC, sigma5SqFrtC,
sigma6Sq, sigma6SqSecC, sigma6SqThdC, sigma6SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu4MinusMu1, diffOfMeansMu4MinusMu1SecC, diffOfMeansMu4MinusMu1ThdC, diffOfMeansMu4MinusMu1FrtC,
diffOfMeansMu5MinusMu1, diffOfMeansMu5MinusMu1SecC, diffOfMeansMu5MinusMu1ThdC, diffOfMeansMu5MinusMu1FrtC,
diffOfMeansMu6MinusMu1, diffOfMeansMu6MinusMu1SecC, diffOfMeansMu6MinusMu1ThdC, diffOfMeansMu6MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfMeansMu4MinusMu2, diffOfMeansMu4MinusMu2SecC, diffOfMeansMu4MinusMu2ThdC, diffOfMeansMu4MinusMu2FrtC,
diffOfMeansMu5MinusMu2, diffOfMeansMu5MinusMu2SecC, diffOfMeansMu5MinusMu2ThdC, diffOfMeansMu5MinusMu2FrtC,
diffOfMeansMu6MinusMu2, diffOfMeansMu6MinusMu2SecC, diffOfMeansMu6MinusMu2ThdC, diffOfMeansMu6MinusMu2FrtC,
diffOfMeansMu4MinusMu3, diffOfMeansMu4MinusMu3SecC, diffOfMeansMu4MinusMu3ThdC, diffOfMeansMu4MinusMu3FrtC,
diffOfMeansMu5MinusMu3, diffOfMeansMu5MinusMu3SecC, diffOfMeansMu5MinusMu3ThdC, diffOfMeansMu5MinusMu3FrtC,
diffOfMeansMu6MinusMu3, diffOfMeansMu6MinusMu3SecC, diffOfMeansMu6MinusMu3ThdC, diffOfMeansMu6MinusMu3FrtC,
diffOfMeansMu5MinusMu4, diffOfMeansMu5MinusMu4SecC, diffOfMeansMu5MinusMu4ThdC, diffOfMeansMu5MinusMu4FrtC,
diffOfMeansMu6MinusMu4, diffOfMeansMu6MinusMu4SecC, diffOfMeansMu6MinusMu4ThdC, diffOfMeansMu6MinusMu4FrtC,
diffOfMeansMu6MinusMu5, diffOfMeansMu6MinusMu5SecC, diffOfMeansMu6MinusMu5ThdC, diffOfMeansMu6MinusMu5FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS4MinusS1, diffOfVariancesS4MinusS1SecC, diffOfVariancesS4MinusS1ThdC, diffOfVariancesS4MinusS1FrtC,
diffOfVariancesS5MinusS1, diffOfVariancesS5MinusS1SecC, diffOfVariancesS5MinusS1ThdC, diffOfVariancesS5MinusS1FrtC,
diffOfVariancesS6MinusS1, diffOfVariancesS6MinusS1SecC, diffOfVariancesS6MinusS1ThdC, diffOfVariancesS6MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
diffOfVariancesS4MinusS2, diffOfVariancesS4MinusS2SecC, diffOfVariancesS4MinusS2ThdC, diffOfVariancesS4MinusS2FrtC,
diffOfVariancesS5MinusS2, diffOfVariancesS5MinusS2SecC, diffOfVariancesS5MinusS2ThdC, diffOfVariancesS5MinusS2FrtC,
diffOfVariancesS6MinusS2, diffOfVariancesS6MinusS2SecC, diffOfVariancesS6MinusS2ThdC, diffOfVariancesS6MinusS2FrtC,
diffOfVariancesS4MinusS3, diffOfVariancesS4MinusS3SecC, diffOfVariancesS4MinusS3ThdC, diffOfVariancesS4MinusS3FrtC,
diffOfVariancesS5MinusS3, diffOfVariancesS5MinusS3SecC, diffOfVariancesS5MinusS3ThdC, diffOfVariancesS5MinusS3FrtC,
diffOfVariancesS6MinusS3, diffOfVariancesS6MinusS3SecC, diffOfVariancesS6MinusS3ThdC, diffOfVariancesS6MinusS3FrtC,
diffOfVariancesS5MinusS4, diffOfVariancesS5MinusS4SecC, diffOfVariancesS5MinusS4ThdC, diffOfVariancesS5MinusS4FrtC,
diffOfVariancesS6MinusS4, diffOfVariancesS6MinusS4SecC, diffOfVariancesS6MinusS4ThdC, diffOfVariancesS6MinusS4FrtC,
diffOfVariancesS6MinusS5, diffOfVariancesS6MinusS5SecC, diffOfVariancesS6MinusS5ThdC, diffOfVariancesS6MinusS5FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize14, effectSize14SecC, effectSize14ThdC, effectSize14FrtC,
effectSize15, effectSize15SecC, effectSize15ThdC, effectSize15FrtC,
effectSize16, effectSize16SecC, effectSize16ThdC, effectSize16FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC,
effectSize24, effectSize24SecC, effectSize24ThdC, effectSize24FrtC,
effectSize25, effectSize25SecC, effectSize25ThdC, effectSize25FrtC,
effectSize26, effectSize26SecC, effectSize26ThdC, effectSize26FrtC,
effectSize34, effectSize34SecC, effectSize34ThdC, effectSize34FrtC,
effectSize35, effectSize35SecC, effectSize35ThdC, effectSize35FrtC,
effectSize36, effectSize36SecC, effectSize36ThdC, effectSize36FrtC,
effectSize45, effectSize45SecC, effectSize45ThdC, effectSize45FrtC,
effectSize46, effectSize46SecC, effectSize46ThdC, effectSize46FrtC,
effectSize56, effectSize56SecC, effectSize56ThdC, effectSize56FrtC)
#pretty table
print(knitr::kable(out))
# return dataframe
return(df)
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && is.null(sixth)) {
Parameter<-c("mu1",
"mu2",
"mu3",
"mu4",
"mu5",
"sigma1",
"sigma2",
"sigma3",
"sigma4",
"sigma5",
"mu2-mu1",
"mu3-mu1",
"mu4-mu1",
"mu5-mu1",
"mu3-mu2",
"mu4-mu2",
"mu5-mu2",
"mu4-mu3",
"mu5-mu3",
"mu5-mu4",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma4-sigma1",
"sigma5-sigma1",
"sigma3-sigma2",
"sigma4-sigma2",
"sigma5-sigma2",
"sigma4-sigma3",
"sigma5-sigma3",
"sigma5-sigma4",
"delta12",
"delta13",
"delta14",
"delta15",
"delta23",
"delta24",
"delta25",
"delta34",
"delta35",
"delta45")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(mu4Quants[1]),
as.numeric(mu5Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(sigma4Quants[1]),
as.numeric(sigma5Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu4MinusMu1Quants[1]),
as.numeric(diffOfMeansMu5MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu2Quants[1]),
as.numeric(diffOfMeansMu5MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu4Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS4MinusS1Quants[1]),
as.numeric(diffOfVariancesS5MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS2Quants[1]),
as.numeric(diffOfVariancesS5MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS4Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize14Quants[1]),
as.numeric(effectSize15Quants[1]),
as.numeric(effectSize23Quants[1]),
as.numeric(effectSize24Quants[1]),
as.numeric(effectSize25Quants[1]),
as.numeric(effectSize34Quants[1]),
as.numeric(effectSize35Quants[1]),
as.numeric(effectSize45Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(mu4),
mean(mu5),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(sigma4Sq),
mean(sigma5Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu4MinusMu1),
mean(diffOfMeansMu5MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfMeansMu4MinusMu2),
mean(diffOfMeansMu5MinusMu2),
mean(diffOfMeansMu4MinusMu3),
mean(diffOfMeansMu5MinusMu3),
mean(diffOfMeansMu5MinusMu4),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS4MinusS1),
mean(diffOfVariancesS5MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(diffOfVariancesS4MinusS2),
mean(diffOfVariancesS5MinusS2),
mean(diffOfVariancesS4MinusS3),
mean(diffOfVariancesS5MinusS3),
mean(diffOfVariancesS5MinusS4),
mean(effectSize12),
mean(effectSize13),
mean(effectSize14),
mean(effectSize15),
mean(effectSize23),
mean(effectSize24),
mean(effectSize25),
mean(effectSize34),
mean(effectSize35),
mean(effectSize45)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(mu4Quants[2]),
as.numeric(mu5Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(sigma4Quants[2]),
as.numeric(sigma5Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu4MinusMu1Quants[2]),
as.numeric(diffOfMeansMu5MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu2Quants[2]),
as.numeric(diffOfMeansMu5MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu4Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS4MinusS1Quants[2]),
as.numeric(diffOfVariancesS5MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS2Quants[2]),
as.numeric(diffOfVariancesS5MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS4Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize14Quants[2]),
as.numeric(effectSize15Quants[2]),
as.numeric(effectSize23Quants[2]),
as.numeric(effectSize24Quants[2]),
as.numeric(effectSize25Quants[2]),
as.numeric(effectSize34Quants[2]),
as.numeric(effectSize35Quants[2]),
as.numeric(effectSize45Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(mu4),
sd(mu5),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(sigma4Sq),
sd(sigma5Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu4MinusMu1),
sd(diffOfMeansMu5MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfMeansMu4MinusMu2),
sd(diffOfMeansMu5MinusMu2),
sd(diffOfMeansMu4MinusMu3),
sd(diffOfMeansMu5MinusMu3),
sd(diffOfMeansMu5MinusMu4),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS4MinusS1),
sd(diffOfVariancesS5MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(diffOfVariancesS4MinusS2),
sd(diffOfVariancesS5MinusS2),
sd(diffOfVariancesS4MinusS3),
sd(diffOfVariancesS5MinusS3),
sd(diffOfVariancesS5MinusS4),
sd(effectSize12),
sd(effectSize13),
sd(effectSize14),
sd(effectSize15),
sd(effectSize23),
sd(effectSize24),
sd(effectSize25),
sd(effectSize34),
sd(effectSize35),
sd(effectSize45)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
# posterior draw dataframe
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
mu4, mu4SecC, mu4ThdC, mu4FrtC,
mu5, mu5SecC, mu5ThdC, mu5FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
sigma4Sq, sigma4SqSecC, sigma4SqThdC, sigma4SqFrtC,
sigma5Sq, sigma5SqSecC, sigma5SqThdC, sigma5SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu4MinusMu1, diffOfMeansMu4MinusMu1SecC, diffOfMeansMu4MinusMu1ThdC, diffOfMeansMu4MinusMu1FrtC,
diffOfMeansMu5MinusMu1, diffOfMeansMu5MinusMu1SecC, diffOfMeansMu5MinusMu1ThdC, diffOfMeansMu5MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfMeansMu4MinusMu2, diffOfMeansMu4MinusMu2SecC, diffOfMeansMu4MinusMu2ThdC, diffOfMeansMu4MinusMu2FrtC,
diffOfMeansMu5MinusMu2, diffOfMeansMu5MinusMu2SecC, diffOfMeansMu5MinusMu2ThdC, diffOfMeansMu5MinusMu2FrtC,
diffOfMeansMu4MinusMu3, diffOfMeansMu4MinusMu3SecC, diffOfMeansMu4MinusMu3ThdC, diffOfMeansMu4MinusMu3FrtC,
diffOfMeansMu5MinusMu3, diffOfMeansMu5MinusMu3SecC, diffOfMeansMu5MinusMu3ThdC, diffOfMeansMu5MinusMu3FrtC,
diffOfMeansMu5MinusMu4, diffOfMeansMu5MinusMu4SecC, diffOfMeansMu5MinusMu4ThdC, diffOfMeansMu5MinusMu4FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS4MinusS1, diffOfVariancesS4MinusS1SecC, diffOfVariancesS4MinusS1ThdC, diffOfVariancesS4MinusS1FrtC,
diffOfVariancesS5MinusS1, diffOfVariancesS5MinusS1SecC, diffOfVariancesS5MinusS1ThdC, diffOfVariancesS5MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
diffOfVariancesS4MinusS2, diffOfVariancesS4MinusS2SecC, diffOfVariancesS4MinusS2ThdC, diffOfVariancesS4MinusS2FrtC,
diffOfVariancesS5MinusS2, diffOfVariancesS5MinusS2SecC, diffOfVariancesS5MinusS2ThdC, diffOfVariancesS5MinusS2FrtC,
diffOfVariancesS4MinusS3, diffOfVariancesS4MinusS3SecC, diffOfVariancesS4MinusS3ThdC, diffOfVariancesS4MinusS3FrtC,
diffOfVariancesS5MinusS3, diffOfVariancesS5MinusS3SecC, diffOfVariancesS5MinusS3ThdC, diffOfVariancesS5MinusS3FrtC,
diffOfVariancesS5MinusS4, diffOfVariancesS5MinusS4SecC, diffOfVariancesS5MinusS4ThdC, diffOfVariancesS5MinusS4FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize14, effectSize14SecC, effectSize14ThdC, effectSize14FrtC,
effectSize15, effectSize15SecC, effectSize15ThdC, effectSize15FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC,
effectSize24, effectSize24SecC, effectSize24ThdC, effectSize24FrtC,
effectSize25, effectSize25SecC, effectSize25ThdC, effectSize25FrtC,
effectSize34, effectSize34SecC, effectSize34ThdC, effectSize34FrtC,
effectSize35, effectSize35SecC, effectSize35ThdC, effectSize35FrtC,
effectSize45, effectSize45SecC, effectSize45ThdC, effectSize45FrtC)
#pretty table
print(knitr::kable(out))
# return posterior draw dataframe
return(df)
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && is.null(fifth) && is.null(sixth)) {
Parameter<-c("mu1",
"mu2",
"mu3",
"mu4",
"sigma1",
"sigma2",
"sigma3",
"sigma4",
"mu2-mu1",
"mu3-mu1",
"mu4-mu1",
"mu3-mu2",
"mu4-mu2",
"mu4-mu3",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma4-sigma1",
"sigma3-sigma2",
"sigma4-sigma2",
"sigma4-sigma3",
"delta12",
"delta13",
"delta14",
"delta23",
"delta24",
"delta34")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(mu4Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(sigma4Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu4MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu3Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS4MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS3Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize14Quants[1]),
as.numeric(effectSize23Quants[1]),
as.numeric(effectSize24Quants[1]),
as.numeric(effectSize34Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(mu4),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(sigma4Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu4MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfMeansMu4MinusMu2),
mean(diffOfMeansMu4MinusMu3),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS4MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(diffOfVariancesS4MinusS2),
mean(diffOfVariancesS4MinusS3),
mean(effectSize12),
mean(effectSize13),
mean(effectSize14),
mean(effectSize23),
mean(effectSize24),
mean(effectSize34)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(mu4Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(sigma4Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu4MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu3Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS4MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS3Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize14Quants[2]),
as.numeric(effectSize23Quants[2]),
as.numeric(effectSize24Quants[2]),
as.numeric(effectSize34Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(mu4),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(sigma4Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu4MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfMeansMu4MinusMu2),
sd(diffOfMeansMu4MinusMu3),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS4MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(diffOfVariancesS4MinusS2),
sd(diffOfVariancesS4MinusS3),
sd(effectSize12),
sd(effectSize13),
sd(effectSize14),
sd(effectSize23),
sd(effectSize24),
sd(effectSize34)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
#pretty table
print(knitr::kable(out))
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
mu4, mu4SecC, mu4ThdC, mu4FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
sigma4Sq, sigma4SqSecC, sigma4SqThdC, sigma4SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu4MinusMu1, diffOfMeansMu4MinusMu1SecC, diffOfMeansMu4MinusMu1ThdC, diffOfMeansMu4MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfMeansMu4MinusMu2, diffOfMeansMu4MinusMu2SecC, diffOfMeansMu4MinusMu2ThdC, diffOfMeansMu4MinusMu2FrtC,
diffOfMeansMu4MinusMu3, diffOfMeansMu4MinusMu3SecC, diffOfMeansMu4MinusMu3ThdC, diffOfMeansMu4MinusMu3FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS4MinusS1, diffOfVariancesS4MinusS1SecC, diffOfVariancesS4MinusS1ThdC, diffOfVariancesS4MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
diffOfVariancesS4MinusS2, diffOfVariancesS4MinusS2SecC, diffOfVariancesS4MinusS2ThdC, diffOfVariancesS4MinusS2FrtC,
diffOfVariancesS4MinusS3, diffOfVariancesS4MinusS3SecC, diffOfVariancesS4MinusS3ThdC, diffOfVariancesS4MinusS3FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize14, effectSize14SecC, effectSize14ThdC, effectSize14FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC,
effectSize24, effectSize24SecC, effectSize24ThdC, effectSize24FrtC,
effectSize34, effectSize34SecC, effectSize34ThdC, effectSize34FrtC)
return(df)
} else if(length(first)>1 && length(second)>1 && length(third)>1 && is.null(fourth) && is.null(fifth) && is.null(sixth)) {
Parameter<-c("mu1",
"mu2",
"mu3",
"sigma1",
"sigma2",
"sigma3",
"mu2-mu1",
"mu3-mu1",
"mu3-mu2",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma3-sigma2",
"delta12",
"delta13",
"delta23")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize23Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(effectSize12),
mean(effectSize13),
mean(effectSize23)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize23Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(effectSize12),
sd(effectSize13),
sd(effectSize23)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
#pretty table
print(knitr::kable(out))
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC)
return(df)
}
}
anovaplot = function(dataframe, type="rope", sd="sd", ci=0.95){
# load parameter chains from dataframe
mu1=dataframe$mu1; mu1SecC=dataframe$mu1SecC; mu1ThdC=dataframe$mu1ThdC; mu1FrtC=dataframe$mu1FrtC;
mu2=dataframe$mu2; mu2SecC=dataframe$mu2SecC; mu2ThdC=dataframe$mu2ThdC; mu2FrtC=dataframe$mu2FrtC;
mu3=dataframe$mu3; mu3SecC=dataframe$mu3SecC; mu3ThdC=dataframe$mu3ThdC; mu3FrtC=dataframe$mu3FrtC;
if(!is.null(dataframe$mu4)){
mu4=dataframe$mu4; mu4SecC=dataframe$mu4SecC; mu4ThdC=dataframe$mu4ThdC; mu4FrtC=dataframe$mu4FrtC;
}
if(!is.null(dataframe$mu5)){
mu5=dataframe$mu5; mu5SecC=dataframe$mu5SecC; mu5ThdC=dataframe$mu5ThdC; mu5FrtC=dataframe$mu5FrtC;
}
if(!is.null(dataframe$mu6)){
mu6=dataframe$mu6; mu6SecC=dataframe$mu6SecC; mu6ThdC=dataframe$mu6ThdC; mu6FrtC=dataframe$mu6FrtC;
}
sigma1Sq=dataframe$sigma1Sq; sigma1SqSecC=dataframe$sigma1SqSecC; sigma1SqThdC=dataframe$sigma1SqThdC; sigma1SqFrtC=dataframe$sigma1SqFrtC;
sigma2Sq=dataframe$sigma2Sq; sigma2SqSecC=dataframe$sigma2SqSecC; sigma2SqThdC=dataframe$sigma2SqThdC; sigma2SqFrtC=dataframe$sigma2SqFrtC;
sigma3Sq=dataframe$sigma3Sq; sigma3SqSecC=dataframe$sigma3SqSecC; sigma3SqThdC=dataframe$sigma3SqThdC; sigma3SqFrtC=dataframe$sigma3SqFrtC;
if(!is.null(dataframe$sigma4Sq)){
sigma4Sq=dataframe$sigma4Sq; sigma4SqSecC=dataframe$sigma4SqSecC; sigma4SqThdC=dataframe$sigma4SqThdC; sigma4SqFrtC=dataframe$sigma4SqFrtC;
}
if(!is.null(dataframe$sigma5Sq)){
sigma5Sq=dataframe$sigma5Sq; sigma5SqSecC=dataframe$sigma5SqSecC; sigma5SqThdC=dataframe$sigma5SqThdC; sigma5SqFrtC=dataframe$sigma5SqFrtC;
}
if(!is.null(dataframe$sigma6Sq)){
sigma6Sq=dataframe$sigma6Sq; sigma6SqSecC=dataframe$sigma6SqSecC; sigma6SqThdC=dataframe$sigma6SqThdC; sigma6SqFrtC=dataframe$sigma6SqFrtC;
}
diffOfMeansMu2MinusMu1=dataframe$diffOfMeansMu2MinusMu1; diffOfMeansMu2MinusMu1SecC=dataframe$diffOfMeansMu2MinusMu1SecC; diffOfMeansMu2MinusMu1ThdC=dataframe$diffOfMeansMu2MinusMu1ThdC; diffOfMeansMu2MinusMu1FrtC=dataframe$diffOfMeansMu2MinusMu1FrtC;
diffOfMeansMu3MinusMu1=dataframe$diffOfMeansMu3MinusMu1; diffOfMeansMu3MinusMu1SecC=dataframe$diffOfMeansMu3MinusMu1SecC; diffOfMeansMu3MinusMu1ThdC=dataframe$diffOfMeansMu3MinusMu1ThdC; diffOfMeansMu3MinusMu1FrtC=dataframe$diffOfMeansMu3MinusMu1FrtC;
diffOfMeansMu3MinusMu2=dataframe$diffOfMeansMu3MinusMu2;diffOfMeansMu3MinusMu2SecC=dataframe$diffOfMeansMu3MinusMu2SecC; diffOfMeansMu3MinusMu2ThdC=dataframe$diffOfMeansMu3MinusMu2ThdC; diffOfMeansMu3MinusMu2FrtC=dataframe$diffOfMeansMu3MinusMu2FrtC;
if(!is.null(dataframe$sigma4Sq)){
diffOfMeansMu4MinusMu1=dataframe$diffOfMeansMu4MinusMu1; diffOfMeansMu4MinusMu1SecC=dataframe$diffOfMeansMu4MinusMu1SecC; diffOfMeansMu4MinusMu1ThdC=dataframe$diffOfMeansMu4MinusMu1ThdC; diffOfMeansMu4MinusMu1FrtC=dataframe$diffOfMeansMu4MinusMu1FrtC;
diffOfMeansMu4MinusMu2=dataframe$diffOfMeansMu4MinusMu2; diffOfMeansMu4MinusMu2SecC=dataframe$diffOfMeansMu4MinusMu2SecC; diffOfMeansMu4MinusMu2ThdC=dataframe$diffOfMeansMu4MinusMu2ThdC; diffOfMeansMu4MinusMu2FrtC=dataframe$diffOfMeansMu4MinusMu2FrtC;
diffOfMeansMu4MinusMu3=dataframe$diffOfMeansMu4MinusMu3; diffOfMeansMu4MinusMu3SecC=dataframe$diffOfMeansMu4MinusMu3SecC; diffOfMeansMu4MinusMu3ThdC=dataframe$diffOfMeansMu4MinusMu3ThdC; diffOfMeansMu4MinusMu3FrtC=dataframe$diffOfMeansMu4MinusMu3FrtC;
}
if(!is.null(dataframe$sigma5Sq)){
diffOfMeansMu5MinusMu1=dataframe$diffOfMeansMu5MinusMu1; diffOfMeansMu5MinusMu1SecC=dataframe$diffOfMeansMu5MinusMu1SecC; diffOfMeansMu5MinusMu1ThdC=dataframe$diffOfMeansMu5MinusMu1ThdC; diffOfMeansMu5MinusMu1FrtC=dataframe$diffOfMeansMu5MinusMu1FrtC;
diffOfMeansMu5MinusMu2=dataframe$diffOfMeansMu5MinusMu2; diffOfMeansMu5MinusMu2SecC=dataframe$diffOfMeansMu5MinusMu2SecC; diffOfMeansMu5MinusMu2ThdC=dataframe$diffOfMeansMu5MinusMu2ThdC; diffOfMeansMu5MinusMu2FrtC=dataframe$diffOfMeansMu5MinusMu2FrtC;
diffOfMeansMu5MinusMu3=dataframe$diffOfMeansMu5MinusMu3; diffOfMeansMu5MinusMu3SecC=dataframe$diffOfMeansMu5MinusMu3SecC; diffOfMeansMu5MinusMu3ThdC=dataframe$diffOfMeansMu5MinusMu3ThdC; diffOfMeansMu5MinusMu3FrtC=dataframe$diffOfMeansMu5MinusMu3FrtC;
diffOfMeansMu5MinusMu4=dataframe$diffOfMeansMu5MinusMu4; diffOfMeansMu5MinusMu4SecC=dataframe$diffOfMeansMu5MinusMu4SecC; diffOfMeansMu5MinusMu4ThdC=dataframe$diffOfMeansMu5MinusMu4ThdC; diffOfMeansMu5MinusMu4FrtC=dataframe$diffOfMeansMu5MinusMu4FrtC;
}
if(!is.null(dataframe$sigma6Sq)){
diffOfMeansMu6MinusMu1=dataframe$diffOfMeansMu6MinusMu1; diffOfMeansMu6MinusMu1SecC=dataframe$diffOfMeansMu6MinusMu1SecC; diffOfMeansMu6MinusMu1ThdC=dataframe$diffOfMeansMu6MinusMu1ThdC; diffOfMeansMu6MinusMu1FrtC=dataframe$diffOfMeansMu6MinusMu1FrtC;
diffOfMeansMu6MinusMu2=dataframe$diffOfMeansMu6MinusMu2; diffOfMeansMu6MinusMu2SecC=dataframe$diffOfMeansMu6MinusMu2SecC; diffOfMeansMu6MinusMu2ThdC=dataframe$diffOfMeansMu6MinusMu2ThdC; diffOfMeansMu6MinusMu2FrtC=dataframe$diffOfMeansMu6MinusMu2FrtC;
diffOfMeansMu6MinusMu3=dataframe$diffOfMeansMu6MinusMu3; diffOfMeansMu6MinusMu3SecC=dataframe$diffOfMeansMu6MinusMu3SecC; diffOfMeansMu6MinusMu3ThdC=dataframe$diffOfMeansMu6MinusMu3ThdC; diffOfMeansMu6MinusMu3FrtC=dataframe$diffOfMeansMu6MinusMu3FrtC;
diffOfMeansMu6MinusMu4=dataframe$diffOfMeansMu6MinusMu4; diffOfMeansMu6MinusMu4SecC=dataframe$diffOfMeansMu6MinusMu4SecC; diffOfMeansMu6MinusMu4ThdC=dataframe$diffOfMeansMu6MinusMu4ThdC; diffOfMeansMu6MinusMu4FrtC=dataframe$diffOfMeansMu6MinusMu4FrtC;
diffOfMeansMu6MinusMu5=dataframe$diffOfMeansMu6MinusMu5; diffOfMeansMu6MinusMu5SecC=dataframe$diffOfMeansMu6MinusMu5SecC; diffOfMeansMu6MinusMu5ThdC=dataframe$diffOfMeansMu6MinusMu5ThdC; diffOfMeansMu6MinusMu5FrtC=dataframe$diffOfMeansMu6MinusMu5FrtC;
}
diffOfVariancesS2MinusS1=dataframe$diffOfVariancesS2MinusS1; diffOfVariancesS2MinusS1SecC=dataframe$diffOfVariancesS2MinusS1SecC; diffOfVariancesS2MinusS1ThdC=dataframe$diffOfVariancesS2MinusS1ThdC; diffOfVariancesS2MinusS1FrtC=dataframe$diffOfVariancesS2MinusS1FrtC;
diffOfVariancesS3MinusS1=dataframe$diffOfVariancesS3MinusS1; diffOfVariancesS3MinusS1SecC=dataframe$diffOfVariancesS3MinusS1SecC; diffOfVariancesS3MinusS1ThdC=dataframe$diffOfVariancesS3MinusS1ThdC; diffOfVariancesS3MinusS1FrtC=dataframe$diffOfVariancesS3MinusS1FrtC;
diffOfVariancesS3MinusS2=dataframe$diffOfVariancesS3MinusS2;diffOfVariancesS3MinusS2SecC=dataframe$diffOfVariancesS3MinusS2SecC; diffOfVariancesS3MinusS2ThdC=dataframe$diffOfVariancesS3MinusS2ThdC; diffOfVariancesS3MinusS2FrtC=dataframe$diffOfVariancesS3MinusS2FrtC;
if(!is.null(dataframe$sigma4Sq)){
diffOfVariancesS4MinusS1=dataframe$diffOfVariancesS4MinusS1; diffOfVariancesS4MinusS1SecC=dataframe$diffOfVariancesS4MinusS1SecC; diffOfVariancesS4MinusS1ThdC=dataframe$diffOfVariancesS4MinusS1ThdC; diffOfVariancesS4MinusS1FrtC=dataframe$diffOfVariancesS4MinusS1FrtC;
diffOfVariancesS4MinusS2=dataframe$diffOfVariancesS4MinusS2; diffOfVariancesS4MinusS2SecC=dataframe$diffOfVariancesS4MinusS2SecC; diffOfVariancesS4MinusS2ThdC=dataframe$diffOfVariancesS4MinusS2ThdC; diffOfVariancesS4MinusS2FrtC=dataframe$diffOfVariancesS4MinusS2FrtC;
diffOfVariancesS4MinusS3=dataframe$diffOfVariancesS4MinusS3; diffOfVariancesS4MinusS3SecC=dataframe$diffOfVariancesS4MinusS3SecC; diffOfVariancesS4MinusS3ThdC=dataframe$diffOfVariancesS4MinusS3ThdC; diffOfVariancesS4MinusS3FrtC=dataframe$diffOfVariancesS4MinusS3FrtC;
}
if(!is.null(dataframe$sigma5Sq)){
diffOfVariancesS5MinusS1=dataframe$diffOfVariancesS5MinusS1; diffOfVariancesS5MinusS1SecC=dataframe$diffOfVariancesS5MinusS1SecC; diffOfVariancesS5MinusS1ThdC=dataframe$diffOfVariancesS5MinusS1ThdC; diffOfVariancesS5MinusS1FrtC=dataframe$diffOfVariancesS5MinusS1FrtC;
diffOfVariancesS5MinusS2=dataframe$diffOfVariancesS5MinusS2; diffOfVariancesS5MinusS2SecC=dataframe$diffOfVariancesS5MinusS2SecC; diffOfVariancesS5MinusS2ThdC=dataframe$diffOfVariancesS5MinusS2ThdC; diffOfVariancesS5MinusS2FrtC=dataframe$diffOfVariancesS5MinusS2FrtC;
diffOfVariancesS5MinusS3=dataframe$diffOfVariancesS5MinusS3; diffOfVariancesS5MinusS3SecC=dataframe$diffOfVariancesS5MinusS3SecC; diffOfVariancesS5MinusS3ThdC=dataframe$diffOfVariancesS5MinusS3ThdC; diffOfVariancesS5MinusS3FrtC=dataframe$diffOfVariancesS5MinusS3FrtC;
diffOfVariancesS5MinusS4=dataframe$diffOfVariancesS5MinusS4; diffOfVariancesS5MinusS4SecC=dataframe$diffOfVariancesS5MinusS4SecC; diffOfVariancesS5MinusS4ThdC=dataframe$diffOfVariancesS5MinusS4ThdC; diffOfVariancesS5MinusS4FrtC=dataframe$diffOfVariancesS5MinusS4FrtC;
}
if(!is.null(dataframe$sigma6Sq)){
diffOfVariancesS6MinusS1=dataframe$diffOfVariancesS6MinusS1; diffOfVariancesS6MinusS1SecC=dataframe$diffOfVariancesS6MinusS1SecC; diffOfVariancesS6MinusS1ThdC=dataframe$diffOfVariancesS6MinusS1ThdC; diffOfVariancesS6MinusS1FrtC=dataframe$diffOfVariancesS6MinusS1FrtC;
diffOfVariancesS6MinusS2=dataframe$diffOfVariancesS6MinusS2; diffOfVariancesS6MinusS2SecC=dataframe$diffOfVariancesS6MinusS2SecC; diffOfVariancesS6MinusS2ThdC=dataframe$diffOfVariancesS6MinusS2ThdC; diffOfVariancesS6MinusS2FrtC=dataframe$diffOfVariancesS6MinusS2FrtC;
diffOfVariancesS6MinusS3=dataframe$diffOfVariancesS6MinusS3; diffOfVariancesS6MinusS3SecC=dataframe$diffOfVariancesS6MinusS3SecC; diffOfVariancesS6MinusS3ThdC=dataframe$diffOfVariancesS6MinusS3ThdC; diffOfVariancesS6MinusS3FrtC=dataframe$diffOfVariancesS6MinusS3FrtC;
diffOfVariancesS6MinusS4=dataframe$diffOfVariancesS6MinusS4; diffOfVariancesS6MinusS4SecC=dataframe$diffOfVariancesS6MinusS4SecC; diffOfVariancesS6MinusS4ThdC=dataframe$diffOfVariancesS6MinusS4ThdC; diffOfVariancesS6MinusS4FrtC=dataframe$diffOfVariancesS6MinusS4FrtC;
diffOfVariancesS6MinusS5=dataframe$diffOfVariancesS6MinusS5; diffOfVariancesS6MinusS5SecC=dataframe$diffOfVariancesS6MinusS5SecC; diffOfVariancesS6MinusS5ThdC=dataframe$diffOfVariancesS6MinusS5ThdC; diffOfVariancesS6MinusS5FrtC=dataframe$diffOfVariancesS6MinusS5FrtC;
}
effectSize12=dataframe$effectSize12; effectSize12SecC=dataframe$effectSize12SecC; effectSize12ThdC=dataframe$effectSize12ThdC; effectSize12FrtC=dataframe$effectSize12FrtC;
effectSize13=dataframe$effectSize13; effectSize13SecC=dataframe$effectSize13SecC; effectSize13ThdC=dataframe$effectSize13ThdC; effectSize13FrtC=dataframe$effectSize13FrtC;
effectSize23=dataframe$effectSize23; effectSize23SecC=dataframe$effectSize23SecC; effectSize23ThdC=dataframe$effectSize23ThdC; effectSize23FrtC=dataframe$effectSize23FrtC;
if(!is.null(dataframe$sigma4Sq)){
effectSize14=dataframe$effectSize14; effectSize14SecC=dataframe$effectSize14SecC; effectSize14ThdC=dataframe$effectSize14ThdC; effectSize14FrtC=dataframe$effectSize14FrtC;
effectSize24=dataframe$effectSize24; effectSize24SecC=dataframe$effectSize24SecC; effectSize24ThdC=dataframe$effectSize24ThdC; effectSize24FrtC=dataframe$effectSize24FrtC;
effectSize34=dataframe$effectSize34; effectSize34SecC=dataframe$effectSize34SecC; effectSize34ThdC=dataframe$effectSize34ThdC; effectSize34FrtC=dataframe$effectSize34FrtC;
}
if(!is.null(dataframe$sigma5Sq)){
effectSize15=dataframe$effectSize15; effectSize15SecC=dataframe$effectSize15SecC; effectSize15ThdC=dataframe$effectSize15ThdC; effectSize15FrtC=dataframe$effectSize15FrtC;
effectSize25=dataframe$effectSize25; effectSize25SecC=dataframe$effectSize25SecC; effectSize25ThdC=dataframe$effectSize25ThdC; effectSize25FrtC=dataframe$effectSize25FrtC;
effectSize35=dataframe$effectSize35; effectSize35SecC=dataframe$effectSize35SecC; effectSize35ThdC=dataframe$effectSize35ThdC; effectSize35FrtC=dataframe$effectSize35FrtC;
effectSize45=dataframe$effectSize45; effectSize45SecC=dataframe$effectSize45SecC; effectSize45ThdC=dataframe$effectSize45ThdC; effectSize45FrtC=dataframe$effectSize45FrtC;
}
if(!is.null(dataframe$sigma6Sq)){
effectSize16=dataframe$effectSize16; effectSize16SecC=dataframe$effectSize16SecC; effectSize16ThdC=dataframe$effectSize16ThdC; effectSize16FrtC=dataframe$effectSize16FrtC;
effectSize26=dataframe$effectSize26; effectSize26SecC=dataframe$effectSize26SecC; effectSize26ThdC=dataframe$effectSize26ThdC; effectSize26FrtC=dataframe$effectSize26FrtC;
effectSize36=dataframe$effectSize36; effectSize36SecC=dataframe$effectSize36SecC; effectSize36ThdC=dataframe$effectSize36ThdC; effectSize36FrtC=dataframe$effectSize36FrtC;
effectSize46=dataframe$effectSize46; effectSize46SecC=dataframe$effectSize46SecC; effectSize46ThdC=dataframe$effectSize46ThdC; effectSize46FrtC=dataframe$effectSize46FrtC;
effectSize56=dataframe$effectSize56; effectSize56SecC=dataframe$effectSize56SecC; effectSize56ThdC=dataframe$effectSize56ThdC; effectSize56FrtC=dataframe$effectSize56FrtC;
}
# posterior plots for each parameter
if(type=="pars"){
# MU1
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu1,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[1]))
plot(mu1,ty="l",col="cornflowerblue",xlab=expression(mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=mu1)
mcmcObj2=coda::mcmc(data=mu1SecC)
mcmcObj3=coda::mcmc(data=mu1ThdC)
mcmcObj4=coda::mcmc(data=mu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu1)
# MU2
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu2,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[2]))
plot(mu2,ty="l",col="cornflowerblue",xlab=expression(mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=mu2)
mcmcObj2=coda::mcmc(data=mu2SecC)
mcmcObj3=coda::mcmc(data=mu2ThdC)
mcmcObj4=coda::mcmc(data=mu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj,auto.layout = FALSE)
stats::acf(mu2)
# MU3
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu3,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[3]))
plot(mu3,ty="l",col="cornflowerblue",xlab=expression(mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=mu3)
mcmcObj2=coda::mcmc(data=mu3SecC)
mcmcObj3=coda::mcmc(data=mu3ThdC)
mcmcObj4=coda::mcmc(data=mu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj,auto.layout = FALSE)
stats::acf(mu3)
# MU4
if(!is.null(dataframe$mu4)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu4,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[4]))
plot(mu4,ty="l",col="cornflowerblue",xlab=expression(mu[4]),ylab="")
mcmcObj1=coda::mcmc(data=mu4)
mcmcObj2=coda::mcmc(data=mu4SecC)
mcmcObj3=coda::mcmc(data=mu4ThdC)
mcmcObj4=coda::mcmc(data=mu4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu4)
}
# MU5
if(!is.null(dataframe$mu5)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu5,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[5]))
plot(mu5,ty="l",col="cornflowerblue",xlab=expression(mu[5]),ylab="")
mcmcObj1=coda::mcmc(data=mu5)
mcmcObj2=coda::mcmc(data=mu5SecC)
mcmcObj3=coda::mcmc(data=mu5ThdC)
mcmcObj4=coda::mcmc(data=mu5FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu5)
}
# MU6
if(!is.null(dataframe$mu6)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu6,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[6]))
plot(mu6,ty="l",col="cornflowerblue",xlab=expression(mu[6]),ylab="")
mcmcObj1=coda::mcmc(data=mu6)
mcmcObj2=coda::mcmc(data=mu6SecC)
mcmcObj3=coda::mcmc(data=mu6ThdC)
mcmcObj4=coda::mcmc(data=mu6FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu6)
}
# SD1
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma1Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[1]^2))
plot(sigma1Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(sigma1Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[1]))
plot(sigma1Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma1Sq)
mcmcObj2=coda::mcmc(data=sigma1SqSecC)
mcmcObj3=coda::mcmc(data=sigma1SqThdC)
mcmcObj4=coda::mcmc(data=sigma1SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma1Sq)
# SD2
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma2Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[2]^2))
plot(sigma2Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(sigma2Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[2]))
plot(sigma2Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma2Sq)
mcmcObj2=coda::mcmc(data=sigma2SqSecC)
mcmcObj3=coda::mcmc(data=sigma2SqThdC)
mcmcObj4=coda::mcmc(data=sigma2SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma2Sq)
# SD3
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma3Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[3]^2))
plot(sigma3Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(sigma3Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[3]))
plot(sigma3Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma3Sq)
mcmcObj2=coda::mcmc(data=sigma3SqSecC)
mcmcObj3=coda::mcmc(data=sigma3SqThdC)
mcmcObj4=coda::mcmc(data=sigma3SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma3Sq)
# SD4
if(!is.null(dataframe$mu4)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma4Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[4]^2))
plot(sigma4Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2),ylab="")
}
if(sd=="sd"){
hist(sigma4Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[4]))
plot(sigma4Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[4]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma4Sq)
mcmcObj2=coda::mcmc(data=sigma4SqSecC)
mcmcObj3=coda::mcmc(data=sigma4SqThdC)
mcmcObj4=coda::mcmc(data=sigma4SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma4Sq)
}
# SD5
if(!is.null(dataframe$mu5)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma5Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[5]^2))
plot(sigma5Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2),ylab="")
}
if(sd=="sd"){
hist(sigma5Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[5]))
plot(sigma5Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[5]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma5Sq)
mcmcObj2=coda::mcmc(data=sigma5SqSecC)
mcmcObj3=coda::mcmc(data=sigma5SqThdC)
mcmcObj4=coda::mcmc(data=sigma5SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma5Sq)
}
# SD6
if(!is.null(dataframe$mu6)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma6Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[6]^2))
plot(sigma6Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2),ylab="")
}
if(sd=="sd"){
hist(sigma6Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[6]))
plot(sigma6Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[6]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma6Sq)
mcmcObj2=coda::mcmc(data=sigma6SqSecC)
mcmcObj3=coda::mcmc(data=sigma6SqThdC)
mcmcObj4=coda::mcmc(data=sigma6SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma6Sq)
}
}
# posterior plot of difference of means & variances
if(type=="diff"){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu2MinusMu1,freq=FALSE,main="Difference of means Group2-Group1",col="cornflowerblue",border="white",xlab=expression(mu[2]-mu[1]))
plot(diffOfMeansMu2MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[2]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu2MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu2MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu2MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu2MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu2MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu3MinusMu2,freq=FALSE,main="Difference of means Group3-Group2",col="cornflowerblue",border="white",xlab=expression(mu[3]-mu[2]))
plot(diffOfMeansMu3MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[3]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu3MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu3MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu3MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu3MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu3MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu3MinusMu1,freq=FALSE,main="Difference of means Group3-Group1",col="cornflowerblue",border="white",xlab=expression(mu[3]-mu[1]))
plot(diffOfMeansMu3MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[3]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu3MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu3MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu3MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu3MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu3MinusMu1)
if(!is.null(dataframe$mu4)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu4MinusMu1,freq=FALSE,main="Difference of means Group4-Group1",col="cornflowerblue",border="white",xlab=expression(mu[4]-mu[1]))
plot(diffOfMeansMu4MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[4]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu4MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu4MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu4MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu4MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu4MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu4MinusMu2,freq=FALSE,main="Difference of means Group2-Group2",col="cornflowerblue",border="white",xlab=expression(mu[4]-mu[2]))
plot(diffOfMeansMu4MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[4]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu4MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu4MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu4MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu4MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu4MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu4MinusMu3,freq=FALSE,main="Difference of means Group4-Group3",col="cornflowerblue",border="white",xlab=expression(mu[4]-mu[3]))
plot(diffOfMeansMu4MinusMu3,ty="l",col="cornflowerblue",xlab=expression(mu[4]-mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu4MinusMu3)
mcmcObj2=coda::mcmc(data=diffOfMeansMu4MinusMu3SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu4MinusMu3ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu4MinusMu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu4MinusMu3)
}
if(!is.null(dataframe$mu5)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu1,freq=FALSE,main="Difference of means Group5-Group1",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[1]))
plot(diffOfMeansMu5MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu2,freq=FALSE,main="Difference of means Group5-Group2",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[2]))
plot(diffOfMeansMu5MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu3,freq=FALSE,main="Difference of means Group5-Group3",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[3]))
plot(diffOfMeansMu5MinusMu3,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu3)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu3SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu3ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu4,freq=FALSE,main="Difference of means Group5-Group4",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[4]))
plot(diffOfMeansMu5MinusMu4,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[4]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu4)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu4SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu4ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu4)
}
if(!is.null(dataframe$mu6)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu1,freq=FALSE,main="Difference of means Group6-Group1",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[1]))
plot(diffOfMeansMu6MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu2,freq=FALSE,main="Difference of means Group6-Group2",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[2]))
plot(diffOfMeansMu6MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu3,freq=FALSE,main="Difference of means Group6-Group3",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[3]))
plot(diffOfMeansMu6MinusMu3,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu3)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu3SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu3ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu4,freq=FALSE,main="Difference of means Group6-Group4",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[4]))
plot(diffOfMeansMu6MinusMu4,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[4]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu4)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu4SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu4ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu4)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu5,freq=FALSE,main="Difference of means Group6-Group5",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[5]))
plot(diffOfMeansMu6MinusMu5,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[5]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu5)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu5SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu5ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu5FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu5)
}
# DIFFERENCES OF VARIANCES / STANDARD DEVIATIONS
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS2MinusS1 ,freq=FALSE,main="Difference of variances Group2-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[2]^2-sigma[1]^2))
plot(diffOfVariancesS2MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[2]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS2MinusS1,freq=FALSE,main="Difference of standard deviations Group2-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[2]-sigma[1]))
plot(diffOfVariancesS2MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[2]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS2MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS2MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS2MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS2MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS2MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS3MinusS2,freq=FALSE,main="Difference of variances Group3-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[3]^2-sigma[2]^2))
plot(diffOfVariancesS3MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[3]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS3MinusS2,freq=FALSE,main="Difference of standard deviations Group3-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[3]-sigma[2]))
plot(diffOfVariancesS3MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[3]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS3MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS3MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS3MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS3MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS3MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS3MinusS1,freq=FALSE,main="Difference of variances Group3-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[3]^2-sigma[1]^2))
plot(diffOfVariancesS3MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[3]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS3MinusS1,freq=FALSE,main="Difference of standard deviations Group3-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[3]-sigma[1]))
plot(diffOfVariancesS3MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[3]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS3MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS3MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS3MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS3MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS3MinusS1)
if(!is.null(dataframe$mu4)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS4MinusS1,freq=FALSE,main="Difference of variances Group4-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[4]^2-sigma[1]^2))
plot(diffOfVariancesS4MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS4MinusS1,freq=FALSE,main="Difference of standard deviations Group4-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[4]-sigma[1]))
plot(diffOfVariancesS4MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[4]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS4MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS4MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS4MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS4MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS4MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS4MinusS2,freq=FALSE,main="Difference of variances Group4-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[4]^2-sigma[2]^2))
plot(diffOfVariancesS4MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS4MinusS2,freq=FALSE,main="Difference of standard deviations Group4-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[4]-sigma[2]))
plot(diffOfVariancesS4MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[4]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS4MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS4MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS4MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS4MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS4MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS4MinusS3,freq=FALSE,main="Difference of variances Group4-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[4]^2-sigma[3]^2))
plot(diffOfVariancesS4MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2-sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS4MinusS3,freq=FALSE,main="Difference of standard deviations Group4-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[4]-sigma[3]))
plot(diffOfVariancesS4MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[4]-sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS4MinusS3)
mcmcObj2=coda::mcmc(data=diffOfVariancesS4MinusS3SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS4MinusS3ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS4MinusS3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS4MinusS3)
}
if(!is.null(dataframe$mu5)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS1,freq=FALSE,main="Difference of variances Group5-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[1]^2))
plot(diffOfVariancesS5MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS1,freq=FALSE,main="Difference of standard deviations Group5-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[1]))
plot(diffOfVariancesS5MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS2,freq=FALSE,main="Difference of variances Group5-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[2]^2))
plot(diffOfVariancesS5MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS2,freq=FALSE,main="Difference of standard deviations Group5-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[2]))
plot(diffOfVariancesS5MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS3,freq=FALSE,main="Difference of variances Group5-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[3]^2))
plot(diffOfVariancesS5MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS3,freq=FALSE,main="Difference of standard deviations Group5-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[3]))
plot(diffOfVariancesS5MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS3)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS3SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS3ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS4,freq=FALSE,main="Difference of variances Group5-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[4]^2))
plot(diffOfVariancesS5MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[4]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS4,freq=FALSE,main="Difference of standard deviations Group5-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[4]))
plot(diffOfVariancesS5MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[4]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS4)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS4SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS4ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS4)
}
if(!is.null(dataframe$mu6)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS1,freq=FALSE,main="Difference of variances Group6-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[1]^2))
plot(diffOfVariancesS6MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS1,freq=FALSE,main="Difference of standard deviations Group6-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[1]))
plot(diffOfVariancesS6MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS2,freq=FALSE,main="Difference of variances Group6-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[2]^2))
plot(diffOfVariancesS6MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS2,freq=FALSE,main="Difference of standard deviations Group6-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[2]))
plot(diffOfVariancesS6MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS3,freq=FALSE,main="Difference of variances Group6-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[3]^2))
plot(diffOfVariancesS6MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS3,freq=FALSE,main="Difference of standard deviations Group6-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[3]))
plot(diffOfVariancesS6MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS3)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS3SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS3ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS4,freq=FALSE,main="Difference of variances Group6-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[4]^2))
plot(diffOfVariancesS6MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[4]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS4,freq=FALSE,main="Difference of standard deviations Group6-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[4]))
plot(diffOfVariancesS6MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[4]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS4)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS4SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS4ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS4)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS5,freq=FALSE,main="Difference of variances Group6-Group5",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[5]^2))
plot(diffOfVariancesS6MinusS5,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[5]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS5,freq=FALSE,main="Difference of standard deviations Group6-Group5",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[5]))
plot(diffOfVariancesS6MinusS5,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[5]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS5)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS5SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS5ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS5FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS5)
}
}
# posterior plot of effect size
if(type=="effect"){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize12,freq=FALSE,main="Effect size Group1-Group2",col="cornflowerblue",border="white",xlab=expression(delta[12]))
plot(effectSize12,ty="l",col="cornflowerblue",xlab=expression(delta[12]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize12)
mcmcObj2=coda::mcmc(data=effectSize12SecC)
mcmcObj3=coda::mcmc(data=effectSize12ThdC)
mcmcObj4=coda::mcmc(data=effectSize12FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize12)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize23,freq=FALSE,main="Effect size Group2-Group3",col="cornflowerblue",border="white",xlab=expression(delta[23]))
plot(effectSize23,ty="l",col="cornflowerblue",xlab=expression(delta[23]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize23)
mcmcObj2=coda::mcmc(data=effectSize23SecC)
mcmcObj3=coda::mcmc(data=effectSize23ThdC)
mcmcObj4=coda::mcmc(data=effectSize23FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize23)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize13,freq=FALSE,main="Effect size Group1-Group3",col="cornflowerblue",border="white",xlab=expression(delta[13]))
plot(effectSize13,ty="l",col="cornflowerblue",xlab=expression(delta[13]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize13)
mcmcObj2=coda::mcmc(data=effectSize13SecC)
mcmcObj3=coda::mcmc(data=effectSize13ThdC)
mcmcObj4=coda::mcmc(data=effectSize13FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize13)
if(!is.null(dataframe$mu4)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize14,freq=FALSE,main="Effect size Group1-Group4",col="cornflowerblue",border="white",xlab=expression(delta[14]))
plot(effectSize14,ty="l",col="cornflowerblue",xlab=expression(delta[14]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize14)
mcmcObj2=coda::mcmc(data=effectSize14SecC)
mcmcObj3=coda::mcmc(data=effectSize14ThdC)
mcmcObj4=coda::mcmc(data=effectSize14FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize14)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize24,freq=FALSE,main="Effect size Group2-Group4",col="cornflowerblue",border="white",xlab=expression(delta[24]))
plot(effectSize24,ty="l",col="cornflowerblue",xlab=expression(delta[24]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize24)
mcmcObj2=coda::mcmc(data=effectSize24SecC)
mcmcObj3=coda::mcmc(data=effectSize24ThdC)
mcmcObj4=coda::mcmc(data=effectSize24FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize24)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize34,freq=FALSE,main="Effect size Group3-Group4",col="cornflowerblue",border="white",xlab=expression(delta[34]))
plot(effectSize34,ty="l",col="cornflowerblue",xlab=expression(delta[34]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize34)
mcmcObj2=coda::mcmc(data=effectSize34SecC)
mcmcObj3=coda::mcmc(data=effectSize34ThdC)
mcmcObj4=coda::mcmc(data=effectSize34FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize34)
}
if(!is.null(dataframe$mu5)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize15,freq=FALSE,main="Effect size Group1-Group5",col="cornflowerblue",border="white",xlab=expression(delta[15]))
plot(effectSize15,ty="l",col="cornflowerblue",xlab=expression(delta[15]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize15)
mcmcObj2=coda::mcmc(data=effectSize15SecC)
mcmcObj3=coda::mcmc(data=effectSize15ThdC)
mcmcObj4=coda::mcmc(data=effectSize15FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize15)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize25,freq=FALSE,main="Effect size Group2-Group5",col="cornflowerblue",border="white",xlab=expression(delta[25]))
plot(effectSize25,ty="l",col="cornflowerblue",xlab=expression(delta[25]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize25)
mcmcObj2=coda::mcmc(data=effectSize25SecC)
mcmcObj3=coda::mcmc(data=effectSize25ThdC)
mcmcObj4=coda::mcmc(data=effectSize25FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize25)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize35,freq=FALSE,main="Effect size Group3-Group5",col="cornflowerblue",border="white",xlab=expression(delta[35]))
plot(effectSize35,ty="l",col="cornflowerblue",xlab=expression(delta[35]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize35)
mcmcObj2=coda::mcmc(data=effectSize35SecC)
mcmcObj3=coda::mcmc(data=effectSize35ThdC)
mcmcObj4=coda::mcmc(data=effectSize35FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize35)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize45,freq=FALSE,main="Effect size Group4-Group5",col="cornflowerblue",border="white",xlab=expression(delta[45]))
plot(effectSize45,ty="l",col="cornflowerblue",xlab=expression(delta[45]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize45)
mcmcObj2=coda::mcmc(data=effectSize45SecC)
mcmcObj3=coda::mcmc(data=effectSize45ThdC)
mcmcObj4=coda::mcmc(data=effectSize45FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize45)
}
if(!is.null(dataframe$mu6)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize16,freq=FALSE,main="Effect size Group1-Group6",col="cornflowerblue",border="white",xlab=expression(delta[16]))
plot(effectSize16,ty="l",col="cornflowerblue",xlab=expression(delta[16]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize16)
mcmcObj2=coda::mcmc(data=effectSize16SecC)
mcmcObj3=coda::mcmc(data=effectSize16ThdC)
mcmcObj4=coda::mcmc(data=effectSize16FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize16)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize26,freq=FALSE,main="Effect size Group2-Group6",col="cornflowerblue",border="white",xlab=expression(delta[26]))
plot(effectSize26,ty="l",col="cornflowerblue",xlab=expression(delta[26]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize26)
mcmcObj2=coda::mcmc(data=effectSize26SecC)
mcmcObj3=coda::mcmc(data=effectSize26ThdC)
mcmcObj4=coda::mcmc(data=effectSize26FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize26)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize36,freq=FALSE,main="Effect size Group3-Group6",col="cornflowerblue",border="white",xlab=expression(delta[36]))
plot(effectSize36,ty="l",col="cornflowerblue",xlab=expression(delta[36]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize36)
mcmcObj2=coda::mcmc(data=effectSize36SecC)
mcmcObj3=coda::mcmc(data=effectSize36ThdC)
mcmcObj4=coda::mcmc(data=effectSize36FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize36)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize46,freq=FALSE,main="Effect size Group4-Group6",col="cornflowerblue",border="white",xlab=expression(delta[46]))
plot(effectSize46,ty="l",col="cornflowerblue",xlab=expression(delta[46]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize46)
mcmcObj2=coda::mcmc(data=effectSize46SecC)
mcmcObj3=coda::mcmc(data=effectSize46ThdC)
mcmcObj4=coda::mcmc(data=effectSize46FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize46)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize56,freq=FALSE,main="Effect size Group5-Group6",col="cornflowerblue",border="white",xlab=expression(delta[56]))
plot(effectSize56,ty="l",col="cornflowerblue",xlab=expression(delta[56]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize56)
mcmcObj2=coda::mcmc(data=effectSize56SecC)
mcmcObj3=coda::mcmc(data=effectSize56ThdC)
mcmcObj4=coda::mcmc(data=effectSize56FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize56)
}
}
# ROPE posterior analysis plot
if(type=="rope"){
# utility function for computing the mode
getmode <- function(v) {
uniqv <- unique(v)
uniqv[which.max(tabulate(match(v, uniqv)))]
}
################# POSTERIOR ROPE ANALYSIS FOR DELTA_12 ######################
# Set 2x3 par
dev.new()
newpar <- par(mfrow=c(2,3),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize12,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[12]),main=mtext("Posterior distribution of Effect size "~delta[12], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize12),getmode(effectSize12)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize12,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize12),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize12),3)),side=3,line=1.15,at=min(effectSize12),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize12),3)),side=3,line=1.15,at=max(effectSize12),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_13 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize13,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[13]),main=mtext("Posterior distribution of Effect size "~delta[13], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize13),getmode(effectSize13)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize13,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize13),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize13),3)),side=3,line=1.15,at=min(effectSize13),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize13),3)),side=3,line=1.15,at=max(effectSize13),col="black",cex=0.8,font=2)
####################################### HISTOGRAM DELTA_23 ###############################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize23,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[23]),main=mtext("Posterior distribution of Effect size "~delta[23], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize23),getmode(effectSize23)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize23,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize23),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize23),3)),side=3,line=1.15,at=min(effectSize23),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize23),3)),side=3,line=1.15,at=max(effectSize23),col="black",cex=0.8,font=2)
######################## POSTERIOR ROPE ANALYSIS DELTA_12 #################################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize12,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize12[effectSize12 >effSizeCI[1] & effectSize12 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[12]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
################################## POSTERIOR ROPE ANALYSIS FOR DELTA_[13] ############################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize13,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize13[effectSize13 >effSizeCI[1] & effectSize13 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[13]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_23 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize23,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize23[effectSize23 >effSizeCI[1] & effectSize23 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[23]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
###################################################### ADDITIONAL EFFECT SIZES FOR FOURTH GROUP #########################################################
if(!is.null(dataframe$mu4)){
# Set 2x3 par
dev.new()
newpar <- par(mfrow=c(2,3),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
########################################### HISTOGRAM DELTA_14 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize14,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[14]),main=mtext("Posterior distribution of Effect size "~delta[14], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize14),getmode(effectSize14)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize14,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize14),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize14),3)),side=3,line=1.15,at=min(effectSize14),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize14),3)),side=3,line=1.15,at=max(effectSize14),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_24 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize24,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[24]),main=mtext("Posterior distribution of Effect size "~delta[24], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize24),getmode(effectSize24)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize24,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-2,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,2),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize24),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize24),3)),side=3,line=1.15,at=min(effectSize24),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize24),3)),side=3,line=1.15,at=max(effectSize24),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_34 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize34,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[34]),main=mtext("Posterior distribution of Effect size "~delta[34], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize34),getmode(effectSize34)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize34,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-2,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,2),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize34),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize34),3)),side=3,line=1.15,at=min(effectSize34),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize34),3)),side=3,line=1.15,at=max(effectSize34),col="black",cex=0.8,font=2)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_14 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize14,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize14[effectSize14 >effSizeCI[1] & effectSize14 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[14]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_24 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize24,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize24[effectSize24 >effSizeCI[1] & effectSize24 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[24]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_34 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize34,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize34[effectSize34 >effSizeCI[1] & effectSize34 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[34]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
}
###################################################### ADDITIONAL EFFECT SIZES FOR FIFTH GROUP #########################################################
if(!is.null(dataframe$mu5)){
# Set 2x4 par
dev.new()
newpar <- par(mfrow=c(2,4),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
########################################### HISTOGRAM DELTA_15 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize15,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[15]),main=mtext("Posterior distribution of Effect size "~delta[15], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize15),getmode(effectSize15)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize15,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize15),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize15),3)),side=3,line=1.15,at=min(effectSize15),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize15),3)),side=3,line=1.15,at=max(effectSize15),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_25 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize25,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[25]),main=mtext("Posterior distribution of Effect size "~delta[25], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize25),getmode(effectSize25)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize25,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize25),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize25),3)),side=3,line=1.15,at=min(effectSize25),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize25),3)),side=3,line=1.15,at=max(effectSize25),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_35 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize35,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[35]),main=mtext("Posterior distribution of Effect size "~delta[35], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize35),getmode(effectSize35)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize35,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize35),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize35),3)),side=3,line=1.15,at=min(effectSize35),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize35),3)),side=3,line=1.15,at=max(effectSize35),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_45 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize45,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[45]),main=mtext("Posterior distribution of Effect size "~delta[45], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize45),getmode(effectSize45)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize45,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize45),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize45),3)),side=3,line=1.15,at=min(effectSize45),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize45),3)),side=3,line=1.15,at=max(effectSize45),col="black",cex=0.8,font=2)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_15 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize15,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize15[effectSize15 >effSizeCI[1] & effectSize15 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[15]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_25 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize25,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize25[effectSize25 >effSizeCI[1] & effectSize25 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[25]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_35 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize35,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize35[effectSize35 >effSizeCI[1] & effectSize35 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[35]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_45 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize45,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize45[effectSize45 >effSizeCI[1] & effectSize45 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[45]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
}
###################################################### ADDITIONAL EFFECT SIZES FOR SIXTH GROUP #########################################################
if(!is.null(dataframe$mu6)){
# Set 2x4 par
dev.new()
newpar <- par(mfrow=c(2,5),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
########################################### HISTOGRAM DELTA_16 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize16,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[16]),main=mtext("Posterior distribution of Effect size "~delta[16], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize16),getmode(effectSize16)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize16,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize16),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize16),3)),side=3,line=1.15,at=min(effectSize16),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize16),3)),side=3,line=1.15,at=max(effectSize16),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_26 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize26,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[26]),main=mtext("Posterior distribution of Effect size "~delta[26], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize26),getmode(effectSize26)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize26,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize26),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize26),3)),side=3,line=1.15,at=min(effectSize26),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize26),3)),side=3,line=1.15,at=max(effectSize26),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_36 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize36,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[36]),main=mtext("Posterior distribution of Effect size "~delta[36], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize36),getmode(effectSize36)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize36,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize36),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize36),3)),side=3,line=1.15,at=min(effectSize36),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize36),3)),side=3,line=1.15,at=max(effectSize36),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_46 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize46,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[46]),main=mtext("Posterior distribution of Effect size "~delta[46], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize46),getmode(effectSize46)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize46,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize46),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize46),3)),side=3,line=1.15,at=min(effectSize46),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize46),3)),side=3,line=1.15,at=max(effectSize46),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_56 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize56,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[56]),main=mtext("Posterior distribution of Effect size "~delta[56], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize56),getmode(effectSize56)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize56,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize56),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize56),3)),side=3,line=1.15,at=min(effectSize56),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize56),3)),side=3,line=1.15,at=max(effectSize56),col="black",cex=0.8,font=2)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_15 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize16,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize16[effectSize16 >effSizeCI[1] & effectSize16 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[16]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_26 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize26,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize26[effectSize26 >effSizeCI[1] & effectSize26 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[26]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_36 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize36,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize36[effectSize36 >effSizeCI[1] & effectSize36 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[36]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_46 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize46,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize46[effectSize46 >effSizeCI[1] & effectSize46 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[46]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_56 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize56,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize56[effectSize56 >effSizeCI[1] & effectSize56 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[56]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
}
}
}
assumption.check = function(x1,x2,x3,x4=NULL,x5=NULL,x6=NULL,conf.level=0.95){
check=TRUE;
if(shapiro.test(x1)$p.value < 1-conf.level | shapiro.test(x2)$p.value < 1-conf.level | shapiro.test(x3)$p.value < 1-conf.level){
check=FALSE;
}
if(!is.null(x4)){
if(shapiro.test(x4)$p.value < 1-conf.level){
check=FALSE;
}
}
if(!is.null(x5)){
if(shapiro.test(x5)$p.value < 1-conf.level){
check=FALSE;
}
}
if(!is.null(x6)){
if(shapiro.test(x6)$p.value < 1-conf.level){
check=FALSE;
}
}
if(check==TRUE){
message("Model assumptions checked. No significant deviations from normality detected. Bayesian ANOVA can be run safely.")
} else {
warning("Model assumption of normally distributed data in each group is violated.\n All results of the Bayesian ANOVA based on a three-component Gaussian mixture could therefore be unreliable and not trustworthy.")
warning("Run further diagnostics (like Quantile-Quantile-plots) to check if the Bayesian ANOVA can be expected to be robust to the violations of normality\n")
}
# Set 2x3 par
dev.new()
newpar <- par(mfrow=c(2,3))
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(x1,freq=FALSE,col="cornflowerblue",border="white",xlab="First group",main="Histogram of first group")
lines(density(x1))
hist(x2,freq=FALSE,col="cornflowerblue",border="white",xlab="Second group",main="Histogram of second group")
lines(density(x2))
hist(x3,freq=FALSE,col="cornflowerblue",border="white",xlab="Third group",main="Histogram of third group")
lines(density(x3))
qqnorm(x1, col="cornflowerblue", main="Q-Q Plot for first group"); qqline(x1, col = "black")
qqnorm(x2, col="cornflowerblue", main="Q-Q Plot for second group"); qqline(x2, col = "black")
qqnorm(x3, col="cornflowerblue", main="Q-Q Plot for third group"); qqline(x3, col = "black")
if(length(x4)>1 && length(x5)>1 && length(x6)>1){
dev.new()
newpar <- par(mfrow=c(2,3))
hist(x4,freq=FALSE,col="cornflowerblue",border="white",xlab="Fourth group",main="Histogram of fourth group")
lines(density(x4))
hist(x5,freq=FALSE,col="cornflowerblue",border="white",xlab="Fifth group",main="Histogram of fifth group")
lines(density(x5))
hist(x6,freq=FALSE,col="cornflowerblue",border="white",xlab="Sixth group",main="Histogram of sixth group")
lines(density(x6))
qqnorm(x4, col="cornflowerblue", main="Q-Q Plot for fourth group"); qqline(x4, col = "black")
qqnorm(x5, col="cornflowerblue", main="Q-Q Plot for fifth group"); qqline(x5, col = "black")
qqnorm(x6, col="cornflowerblue", main="Q-Q Plot for sixth group"); qqline(x6, col = "black")
} else if(length(x4)>1 && length(x5)>1 && is.null(x6)){
dev.new()
newpar <- par(mfrow=c(2,2))
hist(x4,freq=FALSE,col="cornflowerblue",border="white",xlab="Fourth group",main="Histogram of fourth group")
lines(density(x4))
hist(x5,freq=FALSE,col="cornflowerblue",border="white",xlab="Fifth group",main="Histogram of fifth group")
lines(density(x5))
qqnorm(x4, col="cornflowerblue", main="Q-Q Plot for fourth group"); qqline(x4, col = "black")
qqnorm(x5, col="cornflowerblue", main="Q-Q Plot for fifth group"); qqline(x5, col = "black")
} else if(length(x4)>1 && is.null(x5) && is.null(x6)){
dev.new()
newpar <- par(mfrow=c(2,1))
hist(x4,freq=FALSE,col="cornflowerblue",border="white",xlab="Fourth group",main="Histogram of fourth group")
lines(density(x4))
qqnorm(x4, col="cornflowerblue", main="Q-Q Plot for fourth group"); qqline(x4, col = "black")
}
}
post.pred.check <- function(anovafit, ngroups, out, reps = 50, eta){
plot(density(out),main="Posterior predictive check",ylim=c(0,max(density(out)$y)*1.5))
for(i in 1:reps){
mu1 = sample(anovafit$mu1,size = 1)
mu2 = sample(anovafit$mu2,size = 1)
mu3 = sample(anovafit$mu3,size = 1)
sigma1Sq = sample(anovafit$sigma1Sq,size = 1)
sigma2Sq = sample(anovafit$sigma2Sq,size = 1)
sigma3Sq = sample(anovafit$sigma3Sq,size = 1)
if(ngroups > 3){
mu4 = sample(anovafit$mu4,size = 1)
sigma4Sq = sample(anovafit$sigma4Sq,size = 1)
}
if(ngroups > 4){
mu5 = sample(anovafit$mu5,size = 1)
sigma5Sq = sample(anovafit$sigma5Sq,size = 1)
}
if(ngroups > 5){
mu6 = sample(anovafit$mu6,size = 1)
sigma6Sq = sample(anovafit$sigma6Sq,size = 1)
}
x = seq(min(out),max(out),length.out = 50)
if(ngroups == 3){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
if(ngroups == 4){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq) +
eta[4]*dnorm(x,mu4,sigma4Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
if(ngroups == 5){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq) +
eta[4]*dnorm(x,mu4,sigma4Sq) +
eta[5]*dnorm(x,mu5,sigma5Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
if(ngroups == 6){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq) +
eta[4]*dnorm(x,mu4,sigma4Sq) +
eta[5]*dnorm(x,mu5,sigma5Sq) +
eta[6]*dnorm(x,mu6,sigma6Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
}
}
Try the bayesanova package in your browser
Any scripts or data that you put into this service are public.
bayesanova documentation built on Oct. 28, 2021, 5:09 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions post.pred.check assumption.check anovaplot bayes.anova
Documented in anovaplot assumption.check bayes.anova post.pred.check
bayes.anova <- function(n=10000,first,second,third,fourth=NULL,fifth=NULL,sixth=NULL,hyperpars="custom",burnin=n/2,sd="sd",q=0.1,ci=0.95){
# utility function for computing the mode
getmode <- function(v) {
uniqv <- unique(v)
uniqv[which.max(tabulate(match(v, uniqv)))]
}
Nsim=n # number of steps performed by the Gibbs sampler
if(length(first)>1){
firstComponent = first # set first component
} else {
warning("Please provide at least data for three groups. Data for first group is missing.")
}
if(length(second)>1){
secondComponent = second # set second component
} else {
warning("Please provide at least data for three groups. Data for second group is missing.")
}
if(length(third)>1){
thirdComponent = third # set second component
} else {
warning("Please provide at least data for three groups. Data for third group is missing.")
}
if(length(fourth)>1){
fourthComponent = fourth # set second component
}
if(length(fifth)>1){
fifthComponent = fifth # set second component
}
if(length(sixth)>1){
sixthComponent = sixth # set second component
}
mu1 <- numeric(Nsim) # arrays for parameters in first chain
mu2 <- numeric(Nsim)
mu3 <- numeric(Nsim)
if(length(fourth)>1){
mu4 <- numeric(Nsim)
}
if(length(fifth)>1){
mu5 <- numeric(Nsim)
}
if(length(sixth)>1){
mu6 <- numeric(Nsim)
}
sigma1Sq <- numeric(Nsim)
sigma2Sq <- numeric(Nsim)
sigma3Sq <- numeric(Nsim)
if(length(fourth)>1){
sigma4Sq <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5Sq <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6Sq <- numeric(Nsim)
}
mu1SecC <- numeric(Nsim) # arrays for parameters in first chain
mu2SecC <- numeric(Nsim)
mu3SecC <- numeric(Nsim)
if(length(fourth)>1){
mu4SecC <- numeric(Nsim)
}
if(length(fifth)>1){
mu5SecC <- numeric(Nsim)
}
if(length(sixth)>1){
mu6SecC <- numeric(Nsim)
}
sigma1SqSecC <- numeric(Nsim)
sigma2SqSecC <- numeric(Nsim)
sigma3SqSecC <- numeric(Nsim)
if(length(fourth)>1){
sigma4SqSecC <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5SqSecC <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6SqSecC <- numeric(Nsim)
}
mu1ThdC <- numeric(Nsim) # arrays for parameters in first chain
mu2ThdC <- numeric(Nsim)
mu3ThdC <- numeric(Nsim)
if(length(fourth)>1){
mu4ThdC <- numeric(Nsim)
}
if(length(fifth)>1){
mu5ThdC <- numeric(Nsim)
}
if(length(sixth)>1){
mu6ThdC <- numeric(Nsim)
}
sigma1SqThdC <- numeric(Nsim)
sigma2SqThdC <- numeric(Nsim)
sigma3SqThdC <- numeric(Nsim)
if(length(fourth)>1){
sigma4SqThdC <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5SqThdC <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6SqThdC <- numeric(Nsim)
}
mu1FrtC <- numeric(Nsim) # arrays for parameters in first chain
mu2FrtC <- numeric(Nsim)
mu3FrtC <- numeric(Nsim)
if(length(fourth)>1){
mu4FrtC <- numeric(Nsim)
}
if(length(fifth)>1){
mu5FrtC <- numeric(Nsim)
}
if(length(sixth)>1){
mu6FrtC <- numeric(Nsim)
}
sigma1SqFrtC <- numeric(Nsim)
sigma2SqFrtC <- numeric(Nsim)
sigma3SqFrtC <- numeric(Nsim)
if(length(fourth)>1){
sigma4SqFrtC <- numeric(Nsim)
}
if(length(fifth)>1){
sigma5SqFrtC <- numeric(Nsim)
}
if(length(sixth)>1){
sigma6SqFrtC <- numeric(Nsim)
}
mu1[1]=mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2[1]=mean(secondComponent)
mu3[1]=mean(thirdComponent)
if(length(fourth)>1){
mu4[1]=mean(fourthComponent)
}
if(length(fifth)>1){
mu5[1]=mean(fifthComponent)
}
if(length(sixth)>1){
mu6[1]=mean(sixthComponent)
}
sigma1Sq[1]=var(firstComponent)
sigma2Sq[1]=var(secondComponent)
sigma3Sq[1]=var(thirdComponent)
if(length(fourth)>1){
sigma4Sq[1]=var(fourthComponent)
}
if(length(fifth)>1){
sigma5Sq[1]=var(fifthComponent)
}
if(length(sixth)>1){
sigma6Sq[1]=var(sixthComponent)
}
mu1SecC[1]=20*mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2SecC[1]=20*mean(secondComponent)
mu3SecC[1]=20*mean(thirdComponent)
if(length(fourth)>1){
mu4SecC[1]=20*mean(fourthComponent)
}
if(length(fifth)>1){
mu5SecC[1]=20*mean(fifthComponent)
}
if(length(sixth)>1){
mu6SecC[1]=20*mean(sixthComponent)
}
sigma1SqSecC[1]=20*var(firstComponent)
sigma2SqSecC[1]=20*var(secondComponent)
sigma3SqSecC[1]=20*var(thirdComponent)
if(length(fourth)>1){
sigma4SqSecC[1]=20*var(fourthComponent)
}
if(length(fifth)>1){
sigma5SqSecC[1]=20*var(fifthComponent)
}
if(length(sixth)>1){
sigma6SqSecC[1]=20*var(sixthComponent)
}
mu1ThdC[1]=1/20*mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2ThdC[1]=1/20*mean(secondComponent)
mu3ThdC[1]=1/20*mean(thirdComponent)
if(length(fourth)>1){
mu4ThdC[1]=1/20*mean(fourthComponent)
}
if(length(fifth)>1){
mu5ThdC[1]=1/20*mean(fifthComponent)
}
if(length(sixth)>1){
mu6ThdC[1]=1/20*mean(sixthComponent)
}
sigma1SqThdC[1]=1/20*var(firstComponent)
sigma2SqThdC[1]=1/20*var(secondComponent)
sigma3SqThdC[1]=1/20*var(thirdComponent)
if(length(fourth)>1){
sigma4SqThdC[1]=1/20*var(fourthComponent)
}
if(length(fifth)>1){
sigma5SqThdC[1]=1/20*var(fifthComponent)
}
if(length(sixth)>1){
sigma6SqThdC[1]=1/20*var(sixthComponent)
}
mu1FrtC[1]=10*mean(firstComponent) # initialize parameters from priors of mu_k and sigma_k^2
mu2FrtC[1]=10*mean(secondComponent)
mu3FrtC[1]=10*mean(thirdComponent)
if(length(fourth)>1){
mu4FrtC[1]=10*mean(fourthComponent)
}
if(length(fifth)>1){
mu5FrtC[1]=10*mean(fifthComponent)
}
if(length(sixth)>1){
mu6FrtC[1]=10*mean(sixthComponent)
}
sigma1SqFrtC[1]=10*var(firstComponent)
sigma2SqFrtC[1]=10*var(secondComponent)
sigma3SqFrtC[1]=10*var(thirdComponent)
if(length(fourth)>1){
sigma4SqFrtC[1]=10*var(fourthComponent)
}
if(length(fifth)>1){
sigma5SqFrtC[1]=10*var(fifthComponent)
}
if(length(sixth)>1){
sigma6SqFrtC[1]=10*var(sixthComponent)
}
# Formula parts
N_1_S=length(firstComponent)
N_2_S=length(secondComponent)
N_3_S=length(thirdComponent)
if(length(fourth)>1){
N_4_S=length(fourthComponent)
}
if(length(fifth)>1){
N_5_S=length(fifthComponent)
}
if(length(sixth)>1){
N_6_S=length(sixthComponent)
}
vary1=var(firstComponent)
vary2=var(secondComponent)
vary3=var(thirdComponent)
if(length(fourth)>1){
vary4=var(fourthComponent)
}
if(length(fifth)>1){
vary5=var(fifthComponent)
}
if(length(sixth)>1){
vary6=var(sixthComponent)
}
bary1=mean(firstComponent)
bary2=mean(secondComponent)
bary3=mean(thirdComponent)
if(length(fourth)>1){
bary4=mean(fourthComponent)
}
if(length(fifth)>1){
bary5=mean(fifthComponent)
}
if(length(sixth)>1){
bary6=mean(sixthComponent)
}
# Set hyperparameters
if(hyperpars=="rafterys"){
if(length(first)>1 && length(second)>1 && length(third)>1){
smple=c(firstComponent,secondComponent,thirdComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent,sixthComponent)
}
b0=mean(smple) # Raftery's hyperparameters
B0=var(smple)/2.6*(max(smple)-min(smple))
#B0=sd(smple)*10
c0=1.28 # Raftery's hyperparameters
C0=0.36*var(smple) # Raftery's hyperparameters
#C0=10*sd(smple)
}
if(hyperpars=="custom"){
#b0=var(smple)*100*((mean(smple)/var(smple))*4) # Raftery's hyperparameters
#B0=var(smple)*100
#c0=1.28 # Raftery's hyperparameters
#C0=var(smple)*2.0
#b0=mean(smple) # Raftery's hyperparameters
#B0=var(smple)/0.8*(max(smple)-min(smple))
#c0=30*var(smple) # Raftery's hyperparameters
#C0=900*var(smple) # Raftery's hyperparameters
#b0=mean(smple) # Raftery's hyperparameters
#B0=100*var(smple)
#c0=50*var(smple) # Raftery's hyperparameters
#C0=900*var(smple) # Raftery's hyperparameters
# Funktionierte bisher ganz gut
if(length(first)>1 && length(second)>1 && length(third)>1){
smple=c(firstComponent,secondComponent,thirdComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent)
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
smple=c(firstComponent,secondComponent,thirdComponent,fourthComponent,fifthComponent,sixthComponent)
}
b0=mean(smple) # Raftery's hyperparameters
B0=250*var(smple)
c0=1.28 # Raftery's hyperparameters
#C0=-0.5*var(smple)+(c0+0.5*(length(firstComp)+length(secondComp)))*5 # Raftery's hyperparameters
if(length(first)>1 && length(second)>1 && length(third)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)))*q
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)+length(fourthComponent)))*q
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)+length(fourthComponent)+length(fifthComponent)))*q
}
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
C0=-0.5*var(smple)+(c0+1/3*(length(firstComponent)+length(secondComponent)+length(thirdComponent)+length(fourthComponent)+length(fifthComponent)+length(sixthComponent)))*q
}
}
# Gibbs sampling via full conditionals
for(t in 2:Nsim){
# FIRST CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
c_1_S=c0+0.5*N_1_S
C_1_S=C0+0.5*sum((firstComponent-mu1[t-1])^2)
sigma1Sq[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
c_2_S=c0+0.5*N_2_S
C_2_S=C0+0.5*sum((secondComponent-mu2[t-1])^2)
sigma2Sq[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
c_3_S=c0+0.5*N_3_S
C_3_S=C0+0.5*sum((thirdComponent-mu3[t-1])^2)
sigma3Sq[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
c_4_S=c0+0.5*N_4_S
C_4_S=C0+0.5*sum((fourthComponent-mu4[t-1])^2)
sigma4Sq[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
c_5_S=c0+0.5*N_5_S
C_5_S=C0+0.5*sum((fifthComponent-mu5[t-1])^2)
sigma5Sq[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
c_6_S=c0+0.5*N_6_S
C_6_S=C0+0.5*sum((sixthComponent-mu6[t-1])^2)
sigma6Sq[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1Sq[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1Sq[t])*N_1_S*bary1+(1/B0)*b0)
mu1[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2Sq[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2Sq[t])*N_2_S*bary2+(1/B0)*b0)
mu2[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3Sq[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3Sq[t])*N_3_S*bary3+(1/B0)*b0)
mu3[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4Sq[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4Sq[t])*N_4_S*bary4+(1/B0)*b0)
mu4[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5Sq[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5Sq[t])*N_5_S*bary5+(1/B0)*b0)
mu5[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6Sq[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6Sq[t])*N_6_S*bary6+(1/B0)*b0)
mu6[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
###############################################################
# SECOND CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
C_1_S=C0+0.5*sum((firstComponent-mu1SecC[t-1])^2)
sigma1SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
C_2_S=C0+0.5*sum((secondComponent-mu2SecC[t-1])^2)
sigma2SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
C_3_S=C0+0.5*sum((thirdComponent-mu3SecC[t-1])^2)
sigma3SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
C_4_S=C0+0.5*sum((fourthComponent-mu4SecC[t-1])^2)
sigma4SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
C_5_S=C0+0.5*sum((fifthComponent-mu5SecC[t-1])^2)
sigma5SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
C_6_S=C0+0.5*sum((sixthComponent-mu6SecC[t-1])^2)
sigma6SqSecC[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1SqSecC[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1SqSecC[t])*N_1_S*bary1+(1/B0)*b0)
mu1SecC[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2SqSecC[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2SqSecC[t])*N_2_S*bary2+(1/B0)*b0)
mu2SecC[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3SqSecC[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3SqSecC[t])*N_3_S*bary3+(1/B0)*b0)
mu3SecC[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4SqSecC[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4SqSecC[t])*N_4_S*bary4+(1/B0)*b0)
mu4SecC[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5SqSecC[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5SqSecC[t])*N_5_S*bary5+(1/B0)*b0)
mu5SecC[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6SqSecC[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6SqSecC[t])*N_6_S*bary6+(1/B0)*b0)
mu6SecC[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
###############################################################
# THIRD CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
C_1_S=C0+0.5*sum((firstComponent-mu1ThdC[t-1])^2)
sigma1SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
C_2_S=C0+0.5*sum((secondComponent-mu2ThdC[t-1])^2)
sigma2SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
C_3_S=C0+0.5*sum((thirdComponent-mu3ThdC[t-1])^2)
sigma3SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
C_4_S=C0+0.5*sum((fourthComponent-mu4ThdC[t-1])^2)
sigma4SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
C_5_S=C0+0.5*sum((fifthComponent-mu5ThdC[t-1])^2)
sigma5SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
C_6_S=C0+0.5*sum((sixthComponent-mu6ThdC[t-1])^2)
sigma6SqThdC[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1SqThdC[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1SqThdC[t])*N_1_S*bary1+(1/B0)*b0)
mu1ThdC[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2SqThdC[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2SqThdC[t])*N_2_S*bary2+(1/B0)*b0)
mu2ThdC[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3SqThdC[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3SqThdC[t])*N_3_S*bary3+(1/B0)*b0)
mu3ThdC[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4SqThdC[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4SqThdC[t])*N_4_S*bary4+(1/B0)*b0)
mu4ThdC[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5SqThdC[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5SqThdC[t])*N_5_S*bary5+(1/B0)*b0)
mu5ThdC[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6SqThdC[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6SqThdC[t])*N_6_S*bary6+(1/B0)*b0)
mu6ThdC[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
###############################################################
# FOURTH CHAIN
# sample sigma_1^2|mu1,mu2,mu3,sigma_2^2,sigma_3^2,S,y
C_1_S=C0+0.5*sum((firstComponent-mu1FrtC[t-1])^2)
sigma1SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_1_S,scale=C_1_S)
# sample sigma_2^2|mu1,mu2,mu3,sigma_1^2,sigma_3^2,S,y
C_2_S=C0+0.5*sum((secondComponent-mu2FrtC[t-1])^2)
sigma2SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_2_S,scale=C_2_S)
# sample sigma_3^2|mu1,mu2,mu3,sigma_1^2,sigma_2^2,S,y
C_3_S=C0+0.5*sum((thirdComponent-mu3FrtC[t-1])^2)
sigma3SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_3_S,scale=C_3_S)
if(length(fourth>1)){
# sample sigma_4^2|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,S,y
C_4_S=C0+0.5*sum((fourthComponent-mu4FrtC[t-1])^2)
sigma4SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_4_S,scale=C_4_S)
}
if(length(fifth>1)){
# sample sigma_5^2|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
C_5_S=C0+0.5*sum((fifthComponent-mu5FrtC[t-1])^2)
sigma5SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_5_S,scale=C_5_S)
}
if(length(sixth>1)){
# sample sigma_6^2|mu1,mu2,mu3,mu4,mu5,mu6,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
C_6_S=C0+0.5*sum((sixthComponent-mu6FrtC[t-1])^2)
sigma6SqFrtC[t]=MCMCpack::rinvgamma(1,shape=c_6_S,scale=C_6_S)
}
# sample mu1|mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_1_S=1/((1/B0)+(1/sigma1SqFrtC[t])*N_1_S) # use updated sigma1Sq[t] here for sigma_1^2
b_1_S=B_1_S*((1/sigma1SqFrtC[t])*N_1_S*bary1+(1/B0)*b0)
mu1FrtC[t]=rnorm(1,mean=b_1_S,sd=B_1_S)
# sample mu2|mu1,mu3,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_2_S=1/((1/B0)+(1/sigma2SqFrtC[t])*N_2_S) # use updated sigma2Sq[t] here for sigma_2^2
b_2_S=B_2_S*((1/sigma2SqFrtC[t])*N_2_S*bary2+(1/B0)*b0)
mu2FrtC[t]=rnorm(1,mean=b_2_S,sd=B_2_S)
# sample mu3|mu1,mu2,sigma_1^2,sigma_2^2,sigma_3^2,S,y
B_3_S=1/((1/B0)+(1/sigma3SqFrtC[t])*N_3_S) # use updated sigma3Sq[t] here for sigma_3^2
b_3_S=B_3_S*((1/sigma3SqFrtC[t])*N_3_S*bary3+(1/B0)*b0)
mu3FrtC[t]=rnorm(1,mean=b_3_S,sd=B_3_S)
if(length(fourth>1)){
# sample mu4|mu1,mu2,mu3,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,S,y
B_4_S=1/((1/B0)+(1/sigma4SqFrtC[t])*N_4_S) # use updated sigma4Sq[t] here for sigma_4^2
b_4_S=B_4_S*((1/sigma4SqFrtC[t])*N_4_S*bary4+(1/B0)*b0)
mu4FrtC[t]=rnorm(1,mean=b_4_S,sd=B_4_S)
}
if(length(fifth>1)){
# sample mu5|mu1,mu2,mu3,mu4,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,S,y
B_5_S=1/((1/B0)+(1/sigma5SqFrtC[t])*N_5_S) # use updated sigma5Sq[t] here for sigma_5^2
b_5_S=B_5_S*((1/sigma5SqFrtC[t])*N_5_S*bary5+(1/B0)*b0)
mu5FrtC[t]=rnorm(1,mean=b_5_S,sd=B_5_S)
}
if(length(sixth>1)){
# sample mu6|mu1,mu2,mu3,mu4,mu5,sigma_1^2,sigma_2^2,sigma_3^2,sigma_4^2,sigma_5^2,sigma_6^2,S,y
B_6_S=1/((1/B0)+(1/sigma6SqFrtC[t])*N_6_S) # use updated sigma6Sq[t] here for sigma_6^2
b_6_S=B_6_S*((1/sigma6SqFrtC[t])*N_6_S*bary6+(1/B0)*b0)
mu6FrtC[t]=rnorm(1,mean=b_6_S,sd=B_6_S)
}
##############################################################
# All chains sample, reiterate
}
# Adapt chains for selected burnin
mu1=mu1[burnin:Nsim]
mu1SecC=mu1SecC[burnin:Nsim]
mu1ThdC=mu1ThdC[burnin:Nsim]
mu1FrtC=mu1FrtC[burnin:Nsim]
mu2=mu2[burnin:Nsim]
mu2SecC=mu2SecC[burnin:Nsim]
mu2ThdC=mu2ThdC[burnin:Nsim]
mu2FrtC=mu2FrtC[burnin:Nsim]
mu3=mu3[burnin:Nsim]
mu3SecC=mu3SecC[burnin:Nsim]
mu3ThdC=mu3ThdC[burnin:Nsim]
mu3FrtC=mu3FrtC[burnin:Nsim]
if(length(fourth)>1){
mu4=mu4[burnin:Nsim]
mu4SecC=mu4SecC[burnin:Nsim]
mu4ThdC=mu4ThdC[burnin:Nsim]
mu4FrtC=mu4FrtC[burnin:Nsim]
}
if(length(fifth)>1){
mu5=mu5[burnin:Nsim]
mu5SecC=mu5SecC[burnin:Nsim]
mu5ThdC=mu5ThdC[burnin:Nsim]
mu5FrtC=mu5FrtC[burnin:Nsim]
}
if(length(sixth)>1){
mu6=mu6[burnin:Nsim]
mu6SecC=mu6SecC[burnin:Nsim]
mu6ThdC=mu6ThdC[burnin:Nsim]
mu6FrtC=mu6FrtC[burnin:Nsim]
}
sigma1Sq=sigma1Sq[burnin:Nsim]
sigma1SqSecC=sigma1SqSecC[burnin:Nsim]
sigma1SqThdC=sigma1SqThdC[burnin:Nsim]
sigma1SqFrtC=sigma1SqFrtC[burnin:Nsim]
sigma2Sq=sigma2Sq[burnin:Nsim]
sigma2SqSecC=sigma2SqSecC[burnin:Nsim]
sigma2SqThdC=sigma2SqThdC[burnin:Nsim]
sigma2SqFrtC=sigma2SqFrtC[burnin:Nsim]
sigma3Sq=sigma3Sq[burnin:Nsim]
sigma3SqSecC=sigma3SqSecC[burnin:Nsim]
sigma3SqThdC=sigma3SqThdC[burnin:Nsim]
sigma3SqFrtC=sigma3SqFrtC[burnin:Nsim]
if(length(fourth)>1){
sigma4Sq=sigma4Sq[burnin:Nsim]
sigma4SqSecC=sigma4SqSecC[burnin:Nsim]
sigma4SqThdC=sigma4SqThdC[burnin:Nsim]
sigma4SqFrtC=sigma4SqFrtC[burnin:Nsim]
}
if(length(fifth)>1){
sigma5Sq=sigma5Sq[burnin:Nsim]
sigma5SqSecC=sigma5SqSecC[burnin:Nsim]
sigma5SqThdC=sigma5SqThdC[burnin:Nsim]
sigma5SqFrtC=sigma5SqFrtC[burnin:Nsim]
}
if(length(sixth)>1){
sigma6Sq=sigma6Sq[burnin:Nsim]
sigma6SqSecC=sigma6SqSecC[burnin:Nsim]
sigma6SqThdC=sigma6SqThdC[burnin:Nsim]
sigma6SqFrtC=sigma6SqFrtC[burnin:Nsim]
}
if(sd=="sd"){
sigma1Sq=sqrt(sigma1Sq)
sigma1SqSecC=sqrt(sigma1SqSecC)
sigma1SqThdC=sqrt(sigma1SqThdC)
sigma1SqFrtC=sqrt(sigma1SqFrtC)
sigma2Sq=sqrt(sigma2Sq)
sigma2SqSecC=sqrt(sigma2SqSecC)
sigma2SqThdC=sqrt(sigma2SqThdC)
sigma2SqFrtC=sqrt(sigma2SqFrtC)
sigma3Sq=sqrt(sigma3Sq)
sigma3SqSecC=sqrt(sigma3SqSecC)
sigma3SqThdC=sqrt(sigma3SqThdC)
sigma3SqFrtC=sqrt(sigma3SqFrtC)
if(length(fourth)>1){
sigma4Sq=sqrt(sigma4Sq)
sigma4SqSecC=sqrt(sigma4SqSecC)
sigma4SqThdC=sqrt(sigma4SqThdC)
sigma4SqFrtC=sqrt(sigma4SqFrtC)
}
if(length(fifth)>1){
sigma5Sq=sqrt(sigma5Sq)
sigma5SqSecC=sqrt(sigma5SqSecC)
sigma5SqThdC=sqrt(sigma5SqThdC)
sigma5SqFrtC=sqrt(sigma5SqFrtC)
}
if(length(sixth)>1){
sigma6Sq=sqrt(sigma6Sq)
sigma6SqSecC=sqrt(sigma6SqSecC)
sigma6SqThdC=sqrt(sigma6SqThdC)
sigma6SqFrtC=sqrt(sigma6SqFrtC)
}
}
diffOfMeansMu2MinusMu1=mu2-mu1
diffOfMeansMu2MinusMu1SecC=mu2SecC-mu1SecC
diffOfMeansMu2MinusMu1ThdC=mu2ThdC-mu1ThdC
diffOfMeansMu2MinusMu1FrtC=mu2FrtC-mu1FrtC
diffOfMeansMu3MinusMu1=mu3-mu1
diffOfMeansMu3MinusMu1SecC=mu3SecC-mu1SecC
diffOfMeansMu3MinusMu1ThdC=mu3ThdC-mu1ThdC
diffOfMeansMu3MinusMu1FrtC=mu3FrtC-mu1FrtC
diffOfMeansMu3MinusMu2=mu3-mu2
diffOfMeansMu3MinusMu2SecC=mu3SecC-mu2SecC
diffOfMeansMu3MinusMu2ThdC=mu3ThdC-mu2ThdC
diffOfMeansMu3MinusMu2FrtC=mu3FrtC-mu2FrtC
if(length(fourth)>1){
diffOfMeansMu4MinusMu1=mu4-mu1
diffOfMeansMu4MinusMu1SecC=mu4SecC-mu1SecC
diffOfMeansMu4MinusMu1ThdC=mu4ThdC-mu1ThdC
diffOfMeansMu4MinusMu1FrtC=mu4FrtC-mu1FrtC
diffOfMeansMu4MinusMu2=mu4-mu2
diffOfMeansMu4MinusMu2SecC=mu4SecC-mu2SecC
diffOfMeansMu4MinusMu2ThdC=mu4ThdC-mu2ThdC
diffOfMeansMu4MinusMu2FrtC=mu4FrtC-mu2FrtC
diffOfMeansMu4MinusMu3=mu4-mu3
diffOfMeansMu4MinusMu3SecC=mu4SecC-mu3SecC
diffOfMeansMu4MinusMu3ThdC=mu4ThdC-mu3ThdC
diffOfMeansMu4MinusMu3FrtC=mu4FrtC-mu3FrtC
}
if(length(fourth)>1 && length(fifth)>1){
diffOfMeansMu5MinusMu1=mu5-mu1
diffOfMeansMu5MinusMu1SecC=mu5SecC-mu1SecC
diffOfMeansMu5MinusMu1ThdC=mu5ThdC-mu1ThdC
diffOfMeansMu5MinusMu1FrtC=mu5FrtC-mu1FrtC
diffOfMeansMu5MinusMu2=mu5-mu2
diffOfMeansMu5MinusMu2SecC=mu5SecC-mu2SecC
diffOfMeansMu5MinusMu2ThdC=mu5ThdC-mu2ThdC
diffOfMeansMu5MinusMu2FrtC=mu5FrtC-mu2FrtC
diffOfMeansMu5MinusMu3=mu5-mu3
diffOfMeansMu5MinusMu3SecC=mu5SecC-mu3SecC
diffOfMeansMu5MinusMu3ThdC=mu5ThdC-mu3ThdC
diffOfMeansMu5MinusMu3FrtC=mu5FrtC-mu3FrtC
diffOfMeansMu5MinusMu4=mu5-mu4
diffOfMeansMu5MinusMu4SecC=mu5SecC-mu4SecC
diffOfMeansMu5MinusMu4ThdC=mu5ThdC-mu4ThdC
diffOfMeansMu5MinusMu4FrtC=mu5FrtC-mu4FrtC
}
if(length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
diffOfMeansMu6MinusMu1=mu6-mu1
diffOfMeansMu6MinusMu1SecC=mu6SecC-mu1SecC
diffOfMeansMu6MinusMu1ThdC=mu6ThdC-mu1ThdC
diffOfMeansMu6MinusMu1FrtC=mu6FrtC-mu1FrtC
diffOfMeansMu6MinusMu2=mu6-mu2
diffOfMeansMu6MinusMu2SecC=mu6SecC-mu2SecC
diffOfMeansMu6MinusMu2ThdC=mu6ThdC-mu2ThdC
diffOfMeansMu6MinusMu2FrtC=mu6FrtC-mu2FrtC
diffOfMeansMu6MinusMu3=mu6-mu3
diffOfMeansMu6MinusMu3SecC=mu6SecC-mu3SecC
diffOfMeansMu6MinusMu3ThdC=mu6ThdC-mu3ThdC
diffOfMeansMu6MinusMu3FrtC=mu6FrtC-mu3FrtC
diffOfMeansMu6MinusMu4=mu6-mu4
diffOfMeansMu6MinusMu4SecC=mu6SecC-mu4SecC
diffOfMeansMu6MinusMu4ThdC=mu6ThdC-mu4ThdC
diffOfMeansMu6MinusMu4FrtC=mu6FrtC-mu4FrtC
diffOfMeansMu6MinusMu5=mu6-mu5
diffOfMeansMu6MinusMu5SecC=mu6SecC-mu5SecC
diffOfMeansMu6MinusMu5ThdC=mu6ThdC-mu5ThdC
diffOfMeansMu6MinusMu5FrtC=mu6FrtC-mu5FrtC
}
diffOfVariancesS2MinusS1=sigma2Sq-sigma1Sq
diffOfVariancesS2MinusS1SecC=sigma2SqSecC-sigma1SqSecC
diffOfVariancesS2MinusS1ThdC=sigma2SqThdC-sigma1SqThdC
diffOfVariancesS2MinusS1FrtC=sigma2SqFrtC-sigma1SqFrtC
diffOfVariancesS3MinusS1=sigma3Sq-sigma1Sq
diffOfVariancesS3MinusS1SecC=sigma3SqSecC-sigma1SqSecC
diffOfVariancesS3MinusS1ThdC=sigma3SqThdC-sigma1SqThdC
diffOfVariancesS3MinusS1FrtC=sigma3SqFrtC-sigma1SqFrtC
diffOfVariancesS3MinusS2=sigma3Sq-sigma2Sq
diffOfVariancesS3MinusS2SecC=sigma3SqSecC-sigma2SqSecC
diffOfVariancesS3MinusS2ThdC=sigma3SqThdC-sigma2SqThdC
diffOfVariancesS3MinusS2FrtC=sigma3SqFrtC-sigma2SqFrtC
if(length(fourth)>1){
diffOfVariancesS4MinusS1=sigma4Sq-sigma1Sq
diffOfVariancesS4MinusS1SecC=sigma4SqSecC-sigma1SqSecC
diffOfVariancesS4MinusS1ThdC=sigma4SqThdC-sigma1SqThdC
diffOfVariancesS4MinusS1FrtC=sigma4SqFrtC-sigma1SqFrtC
diffOfVariancesS4MinusS2=sigma4Sq-sigma2Sq
diffOfVariancesS4MinusS2SecC=sigma4SqSecC-sigma2SqSecC
diffOfVariancesS4MinusS2ThdC=sigma4SqThdC-sigma2SqThdC
diffOfVariancesS4MinusS2FrtC=sigma4SqFrtC-sigma2SqFrtC
diffOfVariancesS4MinusS3=sigma4Sq-sigma3Sq
diffOfVariancesS4MinusS3SecC=sigma4SqSecC-sigma3SqSecC
diffOfVariancesS4MinusS3ThdC=sigma4SqThdC-sigma3SqThdC
diffOfVariancesS4MinusS3FrtC=sigma4SqFrtC-sigma3SqFrtC
}
if(length(fourth)>1 && length(fifth)>1){
diffOfVariancesS5MinusS1=sigma5Sq-sigma1Sq
diffOfVariancesS5MinusS1SecC=sigma5SqSecC-sigma1SqSecC
diffOfVariancesS5MinusS1ThdC=sigma5SqThdC-sigma1SqThdC
diffOfVariancesS5MinusS1FrtC=sigma5SqFrtC-sigma1SqFrtC
diffOfVariancesS5MinusS2=sigma5Sq-sigma2Sq
diffOfVariancesS5MinusS2SecC=sigma5SqSecC-sigma2SqSecC
diffOfVariancesS5MinusS2ThdC=sigma5SqThdC-sigma2SqThdC
diffOfVariancesS5MinusS2FrtC=sigma5SqFrtC-sigma2SqFrtC
diffOfVariancesS5MinusS3=sigma5Sq-sigma3Sq
diffOfVariancesS5MinusS3SecC=sigma5SqSecC-sigma3SqSecC
diffOfVariancesS5MinusS3ThdC=sigma5SqThdC-sigma3SqThdC
diffOfVariancesS5MinusS3FrtC=sigma5SqFrtC-sigma3SqFrtC
diffOfVariancesS5MinusS4=sigma5Sq-sigma4Sq
diffOfVariancesS5MinusS4SecC=sigma5SqSecC-sigma4SqSecC
diffOfVariancesS5MinusS4ThdC=sigma5SqThdC-sigma4SqThdC
diffOfVariancesS5MinusS4FrtC=sigma5SqFrtC-sigma4SqFrtC
}
if(length(fourth)>1 && length(fifth)>1 && length(sixth)>1){
diffOfVariancesS6MinusS1=sigma6Sq-sigma1Sq
diffOfVariancesS6MinusS1SecC=sigma6SqSecC-sigma1SqSecC
diffOfVariancesS6MinusS1ThdC=sigma6SqThdC-sigma1SqThdC
diffOfVariancesS6MinusS1FrtC=sigma6SqFrtC-sigma1SqFrtC
diffOfVariancesS6MinusS2=sigma6Sq-sigma2Sq
diffOfVariancesS6MinusS2SecC=sigma6SqSecC-sigma2SqSecC
diffOfVariancesS6MinusS2ThdC=sigma6SqThdC-sigma2SqThdC
diffOfVariancesS6MinusS2FrtC=sigma6SqFrtC-sigma2SqFrtC
diffOfVariancesS6MinusS3=sigma6Sq-sigma3Sq
diffOfVariancesS6MinusS3SecC=sigma6SqSecC-sigma3SqSecC
diffOfVariancesS6MinusS3ThdC=sigma6SqThdC-sigma3SqThdC
diffOfVariancesS6MinusS3FrtC=sigma6SqFrtC-sigma3SqFrtC
diffOfVariancesS6MinusS4=sigma6Sq-sigma4Sq
diffOfVariancesS6MinusS4SecC=sigma6SqSecC-sigma4SqSecC
diffOfVariancesS6MinusS4ThdC=sigma6SqThdC-sigma4SqThdC
diffOfVariancesS6MinusS4FrtC=sigma6SqFrtC-sigma4SqFrtC
diffOfVariancesS6MinusS5=sigma6Sq-sigma5Sq
diffOfVariancesS6MinusS5SecC=sigma6SqSecC-sigma5SqSecC
diffOfVariancesS6MinusS5ThdC=sigma6SqThdC-sigma5SqThdC
diffOfVariancesS6MinusS5FrtC=sigma6SqFrtC-sigma5SqFrtC
}
# Effect sizes
if(sd=="var"){
effectSize21=(mu1-mu2)/(sqrt(((N_1_S-1)*sigma1Sq+(N_2_S-1)*sigma2Sq)/(N_1_S+N_2_S-2)))
effectSize21SecC=(mu1SecC-mu2SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_2_S-1)*sigma2SqSecC)/(N_1_S+N_2_S-2)))
effectSize21ThdC=(mu1ThdC-mu2ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_2_S-1)*sigma2SqThdC)/(N_1_S+N_2_S-2)))
effectSize21FrtC=(mu1FrtC-mu2FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_2_S-1)*sigma2SqFrtC)/(N_1_S+N_2_S-2)))
effectSize13=(mu1-mu3)/(sqrt(((N_1_S-1)*sigma1Sq+(N_3_S-1)*sigma3Sq)/(N_1_S+N_3_S-2)))
effectSize13SecC=(mu1SecC-mu3SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_1_S+N_3_S-2)))
effectSize13ThdC=(mu1ThdC-mu3ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_1_S+N_3_S-2)))
effectSize13FrtC=(mu1FrtC-mu3FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_1_S+N_3_S-2)))
effectSize23=(mu2-mu3)/(sqrt(((N_2_S-1)*sigma2Sq+(N_3_S-1)*sigma3Sq)/(N_2_S+N_3_S-2)))
effectSize23SecC=(mu2SecC-mu3SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_2_S+N_3_S-2)))
effectSize23ThdC=(mu2ThdC-mu3ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_2_S+N_3_S-2)))
effectSize23FrtC=(mu2FrtC-mu3FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_2_S+N_3_S-2)))
if(length(fourth>1)){
effectSize14=(mu1-mu4)/(sqrt(((N_1_S-1)*sigma1Sq+(N_4_S-1)*sigma4Sq)/(N_1_S+N_4_S-2)))
effectSize14SecC=(mu1SecC-mu4SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_1_S+N_4_S-2)))
effectSize14ThdC=(mu1ThdC-mu4ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_1_S+N_4_S-2)))
effectSize14FrtC=(mu1FrtC-mu4FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_1_S+N_4_S-2)))
effectSize24=(mu2-mu4)/(sqrt(((N_2_S-1)*sigma2Sq+(N_4_S-1)*sigma4Sq)/(N_2_S+N_4_S-2)))
effectSize24SecC=(mu2SecC-mu4SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_2_S+N_4_S-2)))
effectSize24ThdC=(mu2ThdC-mu4ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_2_S+N_4_S-2)))
effectSize24FrtC=(mu2FrtC-mu4FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_2_S+N_4_S-2)))
effectSize34=(mu3-mu4)/(sqrt(((N_3_S-1)*sigma3Sq+(N_4_S-1)*sigma4Sq)/(N_3_S+N_4_S-2)))
effectSize34SecC=(mu3SecC-mu4SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_3_S+N_4_S-2)))
effectSize34ThdC=(mu3ThdC-mu4ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_3_S+N_4_S-2)))
effectSize34FrtC=(mu3FrtC-mu4FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_3_S+N_4_S-2)))
}
if(length(fourth>1) && length(fifth)>1){
effectSize15=(mu1-mu5)/(sqrt(((N_1_S-1)*sigma1Sq+(N_5_S-1)*sigma5Sq)/(N_1_S+N_5_S-2)))
effectSize15SecC=(mu1SecC-mu5SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_1_S+N_5_S-2)))
effectSize15ThdC=(mu1ThdC-mu5ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_1_S+N_5_S-2)))
effectSize15FrtC=(mu1FrtC-mu5FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_1_S+N_5_S-2)))
effectSize25=(mu2-mu5)/(sqrt(((N_2_S-1)*sigma2Sq+(N_5_S-1)*sigma5Sq)/(N_2_S+N_5_S-2)))
effectSize25SecC=(mu2SecC-mu5SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_2_S+N_5_S-2)))
effectSize25ThdC=(mu2ThdC-mu5ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_2_S+N_5_S-2)))
effectSize25FrtC=(mu2FrtC-mu5FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_2_S+N_5_S-2)))
effectSize35=(mu3-mu5)/(sqrt(((N_3_S-1)*sigma3Sq+(N_5_S-1)*sigma5Sq)/(N_3_S+N_5_S-2)))
effectSize35SecC=(mu3SecC-mu5SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_3_S+N_5_S-2)))
effectSize35ThdC=(mu3ThdC-mu5ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_3_S+N_5_S-2)))
effectSize35FrtC=(mu3FrtC-mu5FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_3_S+N_5_S-2)))
effectSize45=(mu4-mu5)/(sqrt(((N_4_S-1)*sigma4Sq+(N_5_S-1)*sigma5Sq)/(N_4_S+N_5_S-2)))
effectSize45SecC=(mu4SecC-mu5SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC+(N_5_S-1)*sigma5SqSecC)/(N_4_S+N_5_S-2)))
effectSize45ThdC=(mu4ThdC-mu5ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC+(N_5_S-1)*sigma5SqThdC)/(N_4_S+N_5_S-2)))
effectSize45FrtC=(mu4FrtC-mu5FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC+(N_5_S-1)*sigma5SqFrtC)/(N_4_S+N_5_S-2)))
}
if(length(fourth>1)&& length(fifth)>1 && length(sixth)>1){
effectSize16=(mu1-mu6)/(sqrt(((N_1_S-1)*sigma1Sq+(N_6_S-1)*sigma6Sq)/(N_1_S+N_6_S-2)))
effectSize16SecC=(mu1SecC-mu6SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_1_S+N_6_S-2)))
effectSize16ThdC=(mu1ThdC-mu6ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_1_S+N_6_S-2)))
effectSize16FrtC=(mu1FrtC-mu6FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_1_S+N_6_S-2)))
effectSize26=(mu2-mu6)/(sqrt(((N_2_S-1)*sigma2Sq+(N_6_S-1)*sigma6Sq)/(N_2_S+N_6_S-2)))
effectSize26SecC=(mu2SecC-mu6SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_2_S+N_6_S-2)))
effectSize26ThdC=(mu2ThdC-mu6ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_2_S+N_6_S-2)))
effectSize26FrtC=(mu2FrtC-mu6FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_2_S+N_6_S-2)))
effectSize36=(mu3-mu6)/(sqrt(((N_3_S-1)*sigma3Sq+(N_6_S-1)*sigma6Sq)/(N_3_S+N_6_S-2)))
effectSize36SecC=(mu3SecC-mu6SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_3_S+N_6_S-2)))
effectSize36ThdC=(mu3ThdC-mu6ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_3_S+N_6_S-2)))
effectSize36FrtC=(mu3FrtC-mu6FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_3_S+N_6_S-2)))
effectSize46=(mu4-mu6)/(sqrt(((N_4_S-1)*sigma4Sq+(N_6_S-1)*sigma6Sq)/(N_4_S+N_6_S-2)))
effectSize46SecC=(mu4SecC-mu6SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_4_S+N_6_S-2)))
effectSize46ThdC=(mu4ThdC-mu6ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_4_S+N_6_S-2)))
effectSize46FrtC=(mu4FrtC-mu6FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_4_S+N_6_S-2)))
effectSize56=(mu5-mu6)/(sqrt(((N_5_S-1)*sigma5Sq+(N_6_S-1)*sigma6Sq)/(N_5_S+N_6_S-2)))
effectSize56SecC=(mu5SecC-mu6SecC)/(sqrt(((N_5_S-1)*sigma5SqSecC+(N_6_S-1)*sigma6SqSecC)/(N_5_S+N_6_S-2)))
effectSize56ThdC=(mu5ThdC-mu6ThdC)/(sqrt(((N_5_S-1)*sigma5SqThdC+(N_6_S-1)*sigma6SqThdC)/(N_5_S+N_6_S-2)))
effectSize56FrtC=(mu5FrtC-mu6FrtC)/(sqrt(((N_5_S-1)*sigma5SqFrtC+(N_6_S-1)*sigma6SqFrtC)/(N_5_S+N_6_S-2)))
}
}
if(sd=="sd"){
effectSize12=(mu1-mu2)/(sqrt(((N_1_S-1)*sigma1Sq+(N_2_S-1)*sigma2Sq)/(N_1_S+N_2_S-2)))
effectSize12SecC=(mu1SecC-mu2SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_2_S-1)*sigma2SqSecC)/(N_1_S+N_2_S-2)))
effectSize12ThdC=(mu1ThdC-mu2ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_2_S-1)*sigma2SqThdC)/(N_1_S+N_2_S-2)))
effectSize12FrtC=(mu1FrtC-mu2FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_2_S-1)*sigma2SqFrtC)/(N_1_S+N_2_S-2)))
effectSize13=(mu1-mu3)/(sqrt(((N_1_S-1)*sigma1Sq+(N_3_S-1)*sigma3Sq)/(N_1_S+N_3_S-2)))
effectSize13SecC=(mu1SecC-mu3SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_1_S+N_3_S-2)))
effectSize13ThdC=(mu1ThdC-mu3ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_1_S+N_3_S-2)))
effectSize13FrtC=(mu1FrtC-mu3FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_1_S+N_3_S-2)))
effectSize23=(mu2-mu3)/(sqrt(((N_2_S-1)*sigma2Sq+(N_3_S-1)*sigma3Sq)/(N_2_S+N_3_S-2)))
effectSize23SecC=(mu2SecC-mu3SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_3_S-1)*sigma3SqSecC)/(N_2_S+N_3_S-2)))
effectSize23ThdC=(mu2ThdC-mu3ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_3_S-1)*sigma3SqThdC)/(N_2_S+N_3_S-2)))
effectSize23FrtC=(mu2FrtC-mu3FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_3_S-1)*sigma3SqFrtC)/(N_2_S+N_3_S-2)))
if(length(fourth>1)){
effectSize14=(mu1-mu4)/(sqrt(((N_1_S-1)*sigma1Sq+(N_4_S-1)*sigma4Sq)/(N_1_S+N_4_S-2)))
effectSize14SecC=(mu1SecC-mu4SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_1_S+N_4_S-2)))
effectSize14ThdC=(mu1ThdC-mu4ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_1_S+N_4_S-2)))
effectSize14FrtC=(mu1FrtC-mu4FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_1_S+N_4_S-2)))
effectSize24=(mu2-mu4)/(sqrt(((N_2_S-1)*sigma2Sq+(N_4_S-1)*sigma4Sq)/(N_2_S+N_4_S-2)))
effectSize24SecC=(mu2SecC-mu4SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_2_S+N_4_S-2)))
effectSize24ThdC=(mu2ThdC-mu4ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_2_S+N_4_S-2)))
effectSize24FrtC=(mu2FrtC-mu4FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_2_S+N_4_S-2)))
effectSize34=(mu3-mu4)/(sqrt(((N_3_S-1)*sigma3Sq+(N_4_S-1)*sigma4Sq)/(N_3_S+N_4_S-2)))
effectSize34SecC=(mu3SecC-mu4SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC+(N_4_S-1)*sigma4SqSecC)/(N_3_S+N_4_S-2)))
effectSize34ThdC=(mu3ThdC-mu4ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC+(N_4_S-1)*sigma4SqThdC)/(N_3_S+N_4_S-2)))
effectSize34FrtC=(mu3FrtC-mu4FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC+(N_4_S-1)*sigma4SqFrtC)/(N_3_S+N_4_S-2)))
}
if(length(fourth>1) && length(fifth)>1){
effectSize15=(mu1-mu5)/(sqrt(((N_1_S-1)*sigma1Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_1_S+N_5_S-2)))
effectSize15SecC=(mu1SecC-mu5SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_1_S+N_5_S-2)))
effectSize15ThdC=(mu1ThdC-mu5ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_1_S+N_5_S-2)))
effectSize15FrtC=(mu1FrtC-mu5FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_1_S+N_5_S-2)))
effectSize25=(mu2-mu5)/(sqrt(((N_2_S-1)*sigma2Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_2_S+N_5_S-2)))
effectSize25SecC=(mu2SecC-mu5SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_2_S+N_5_S-2)))
effectSize25ThdC=(mu2ThdC-mu5ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_2_S+N_5_S-2)))
effectSize25FrtC=(mu2FrtC-mu5FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_2_S+N_5_S-2)))
effectSize35=(mu3-mu5)/(sqrt(((N_3_S-1)*sigma3Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_3_S+N_5_S-2)))
effectSize35SecC=(mu3SecC-mu5SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_3_S+N_5_S-2)))
effectSize35ThdC=(mu3ThdC-mu5ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_3_S+N_5_S-2)))
effectSize35FrtC=(mu3FrtC-mu5FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_3_S+N_5_S-2)))
effectSize45=(mu4-mu5)/(sqrt(((N_4_S-1)*sigma4Sq^2+(N_5_S-1)*sigma5Sq^2)/(N_4_S+N_5_S-2)))
effectSize45SecC=(mu4SecC-mu5SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC^2+(N_5_S-1)*sigma5SqSecC^2)/(N_4_S+N_5_S-2)))
effectSize45ThdC=(mu4ThdC-mu5ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC^2+(N_5_S-1)*sigma5SqThdC^2)/(N_4_S+N_5_S-2)))
effectSize45FrtC=(mu4FrtC-mu5FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC^2+(N_5_S-1)*sigma5SqFrtC^2)/(N_4_S+N_5_S-2)))
}
if(length(fourth>1)&& length(fifth)>1 && length(sixth)>1){
effectSize16=(mu1-mu6)/(sqrt(((N_1_S-1)*sigma1Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_1_S+N_6_S-2)))
effectSize16SecC=(mu1SecC-mu6SecC)/(sqrt(((N_1_S-1)*sigma1SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_1_S+N_6_S-2)))
effectSize16ThdC=(mu1ThdC-mu6ThdC)/(sqrt(((N_1_S-1)*sigma1SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_1_S+N_6_S-2)))
effectSize16FrtC=(mu1FrtC-mu6FrtC)/(sqrt(((N_1_S-1)*sigma1SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_1_S+N_6_S-2)))
effectSize26=(mu2-mu6)/(sqrt(((N_2_S-1)*sigma2Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_2_S+N_6_S-2)))
effectSize26SecC=(mu2SecC-mu6SecC)/(sqrt(((N_2_S-1)*sigma2SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_2_S+N_6_S-2)))
effectSize26ThdC=(mu2ThdC-mu6ThdC)/(sqrt(((N_2_S-1)*sigma2SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_2_S+N_6_S-2)))
effectSize26FrtC=(mu2FrtC-mu6FrtC)/(sqrt(((N_2_S-1)*sigma2SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_2_S+N_6_S-2)))
effectSize36=(mu3-mu6)/(sqrt(((N_3_S-1)*sigma3Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_3_S+N_6_S-2)))
effectSize36SecC=(mu3SecC-mu6SecC)/(sqrt(((N_3_S-1)*sigma3SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_3_S+N_6_S-2)))
effectSize36ThdC=(mu3ThdC-mu6ThdC)/(sqrt(((N_3_S-1)*sigma3SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_3_S+N_6_S-2)))
effectSize36FrtC=(mu3FrtC-mu6FrtC)/(sqrt(((N_3_S-1)*sigma3SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_3_S+N_6_S-2)))
effectSize46=(mu4-mu6)/(sqrt(((N_4_S-1)*sigma4Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_4_S+N_6_S-2)))
effectSize46SecC=(mu4SecC-mu6SecC)/(sqrt(((N_4_S-1)*sigma4SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_4_S+N_6_S-2)))
effectSize46ThdC=(mu4ThdC-mu6ThdC)/(sqrt(((N_4_S-1)*sigma4SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_4_S+N_6_S-2)))
effectSize46FrtC=(mu4FrtC-mu6FrtC)/(sqrt(((N_4_S-1)*sigma4SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_4_S+N_6_S-2)))
effectSize56=(mu5-mu6)/(sqrt(((N_5_S-1)*sigma5Sq^2+(N_6_S-1)*sigma6Sq^2)/(N_5_S+N_6_S-2)))
effectSize56SecC=(mu5SecC-mu6SecC)/(sqrt(((N_5_S-1)*sigma5SqSecC^2+(N_6_S-1)*sigma6SqSecC^2)/(N_5_S+N_6_S-2)))
effectSize56ThdC=(mu5ThdC-mu6ThdC)/(sqrt(((N_5_S-1)*sigma5SqThdC^2+(N_6_S-1)*sigma6SqThdC^2)/(N_5_S+N_6_S-2)))
effectSize56FrtC=(mu5FrtC-mu6FrtC)/(sqrt(((N_5_S-1)*sigma5SqFrtC^2+(N_6_S-1)*sigma6SqFrtC^2)/(N_5_S+N_6_S-2)))
}
}
##################################### CONSOLE OUTPUT ############################################
message("Bayesian ANOVA output:")
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1) {
message("Details: Gaussian-mixture model with six components")
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1) {
message("Details: Gaussian-mixture model with five components")
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1) {
message("Details: Gaussian-mixture model with four components")
} else if(length(first)>1 && length(second)>1 && length(third)>1) {
message("Details: Gaussian-mixture model with three components")
}
# Quantiles
mu1Quants=quantile(mu1,probs=c((1-ci)/2,ci+(1-ci)/2))
mu2Quants=quantile(mu2,probs=c((1-ci)/2,ci+(1-ci)/2))
mu3Quants=quantile(mu3,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
mu4Quants=quantile(mu4,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
mu5Quants=quantile(mu5,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
mu6Quants=quantile(mu6,probs=c((1-ci)/2,ci+(1-ci)/2))
}
sigma1Quants=quantile(sigma1Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
sigma2Quants=quantile(sigma2Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
sigma3Quants=quantile(sigma3Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
sigma4Quants=quantile(sigma4Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
sigma5Quants=quantile(sigma5Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
sigma6Quants=quantile(sigma6Sq,probs=c((1-ci)/2,ci+(1-ci)/2))
}
diffOfMeansMu2MinusMu1Quants=quantile(diffOfMeansMu2MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu3MinusMu1Quants=quantile(diffOfMeansMu3MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu3MinusMu2Quants=quantile(diffOfMeansMu3MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
diffOfMeansMu4MinusMu1Quants=quantile(diffOfMeansMu4MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu4MinusMu2Quants=quantile(diffOfMeansMu4MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu4MinusMu3Quants=quantile(diffOfMeansMu4MinusMu3,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
diffOfMeansMu5MinusMu1Quants=quantile(diffOfMeansMu5MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu5MinusMu2Quants=quantile(diffOfMeansMu5MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu5MinusMu3Quants=quantile(diffOfMeansMu5MinusMu3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu5MinusMu4Quants=quantile(diffOfMeansMu5MinusMu4,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
diffOfMeansMu6MinusMu1Quants=quantile(diffOfMeansMu6MinusMu1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu2Quants=quantile(diffOfMeansMu6MinusMu2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu3Quants=quantile(diffOfMeansMu6MinusMu3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu4Quants=quantile(diffOfMeansMu6MinusMu4,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfMeansMu6MinusMu5Quants=quantile(diffOfMeansMu6MinusMu5,probs=c((1-ci)/2,ci+(1-ci)/2))
}
diffOfVariancesS2MinusS1Quants=quantile(diffOfVariancesS2MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS3MinusS1Quants=quantile(diffOfVariancesS3MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS3MinusS2Quants=quantile(diffOfVariancesS3MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
diffOfVariancesS4MinusS1Quants=quantile(diffOfVariancesS4MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS4MinusS2Quants=quantile(diffOfVariancesS4MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS4MinusS3Quants=quantile(diffOfVariancesS4MinusS3,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
diffOfVariancesS5MinusS1Quants=quantile(diffOfVariancesS5MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS5MinusS2Quants=quantile(diffOfVariancesS5MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS5MinusS3Quants=quantile(diffOfVariancesS5MinusS3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS5MinusS4Quants=quantile(diffOfVariancesS5MinusS4,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
diffOfVariancesS6MinusS1Quants=quantile(diffOfVariancesS6MinusS1,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS2Quants=quantile(diffOfVariancesS6MinusS2,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS3Quants=quantile(diffOfVariancesS6MinusS3,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS4Quants=quantile(diffOfVariancesS6MinusS4,probs=c((1-ci)/2,ci+(1-ci)/2))
diffOfVariancesS6MinusS5Quants=quantile(diffOfVariancesS6MinusS5,probs=c((1-ci)/2,ci+(1-ci)/2))
}
effectSize12Quants=quantile(effectSize12,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize13Quants=quantile(effectSize13,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize23Quants=quantile(effectSize23,probs=c((1-ci)/2,ci+(1-ci)/2))
if(length(fourth>1)){
effectSize14Quants=quantile(effectSize14,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize24Quants=quantile(effectSize24,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize34Quants=quantile(effectSize34,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(fifth>1)){
effectSize15Quants=quantile(effectSize15,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize25Quants=quantile(effectSize25,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize35Quants=quantile(effectSize35,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize45Quants=quantile(effectSize45,probs=c((1-ci)/2,ci+(1-ci)/2))
}
if(length(sixth>1)){
effectSize16Quants=quantile(effectSize16,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize26Quants=quantile(effectSize26,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize36Quants=quantile(effectSize36,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize46Quants=quantile(effectSize46,probs=c((1-ci)/2,ci+(1-ci)/2))
effectSize56Quants=quantile(effectSize56,probs=c((1-ci)/2,ci+(1-ci)/2))
}
# Console Output
if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && length(sixth)>1) {
Parameter<-c("mu1",
"mu2",
"mu3",
"mu4",
"mu5",
"mu6",
"sigma1",
"sigma2",
"sigma3",
"sigma4",
"sigma5",
"sigma6",
"mu2-mu1",
"mu3-mu1",
"mu4-mu1",
"mu5-mu1",
"mu6-mu1",
"mu3-mu2",
"mu4-mu2",
"mu5-mu2",
"mu6-mu2",
"mu4-mu3",
"mu5-mu3",
"mu6-mu3",
"mu5-mu4",
"mu6-mu4",
"mu6-mu5",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma4-sigma1",
"sigma5-sigma1",
"sigma6-sigma1",
"sigma3-sigma2",
"sigma4-sigma2",
"sigma5-sigma2",
"sigma6-sigma2",
"sigma4-sigma3",
"sigma5-sigma3",
"sigma6-sigma3",
"sigma5-sigma4",
"sigma6-sigma4",
"sigma6-sigma5",
"delta12",
"delta13",
"delta14",
"delta15",
"delta16",
"delta23",
"delta24",
"delta25",
"delta26",
"delta34",
"delta35",
"delta36",
"delta45",
"delta46",
"delta56")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(mu4Quants[1]),
as.numeric(mu5Quants[1]),
as.numeric(mu6Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(sigma4Quants[1]),
as.numeric(sigma5Quants[1]),
as.numeric(sigma6Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu4MinusMu1Quants[1]),
as.numeric(diffOfMeansMu5MinusMu1Quants[1]),
as.numeric(diffOfMeansMu6MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu2Quants[1]),
as.numeric(diffOfMeansMu5MinusMu2Quants[1]),
as.numeric(diffOfMeansMu6MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu3Quants[1]),
as.numeric(diffOfMeansMu6MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu4Quants[1]),
as.numeric(diffOfMeansMu6MinusMu4Quants[1]),
as.numeric(diffOfMeansMu6MinusMu5Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS4MinusS1Quants[1]),
as.numeric(diffOfVariancesS5MinusS1Quants[1]),
as.numeric(diffOfVariancesS6MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS2Quants[1]),
as.numeric(diffOfVariancesS5MinusS2Quants[1]),
as.numeric(diffOfVariancesS6MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS3Quants[1]),
as.numeric(diffOfVariancesS6MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS4Quants[1]),
as.numeric(diffOfVariancesS6MinusS4Quants[1]),
as.numeric(diffOfVariancesS6MinusS5Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize14Quants[1]),
as.numeric(effectSize15Quants[1]),
as.numeric(effectSize16Quants[1]),
as.numeric(effectSize23Quants[1]),
as.numeric(effectSize24Quants[1]),
as.numeric(effectSize25Quants[1]),
as.numeric(effectSize26Quants[1]),
as.numeric(effectSize34Quants[1]),
as.numeric(effectSize35Quants[1]),
as.numeric(effectSize36Quants[1]),
as.numeric(effectSize45Quants[1]),
as.numeric(effectSize46Quants[1]),
as.numeric(effectSize56Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(mu4),
mean(mu5),
mean(mu6),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(sigma4Sq),
mean(sigma5Sq),
mean(sigma6Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu4MinusMu1),
mean(diffOfMeansMu5MinusMu1),
mean(diffOfMeansMu6MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfMeansMu4MinusMu2),
mean(diffOfMeansMu5MinusMu2),
mean(diffOfMeansMu6MinusMu2),
mean(diffOfMeansMu4MinusMu3),
mean(diffOfMeansMu5MinusMu3),
mean(diffOfMeansMu6MinusMu3),
mean(diffOfMeansMu5MinusMu4),
mean(diffOfMeansMu6MinusMu4),
mean(diffOfMeansMu6MinusMu5),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS4MinusS1),
mean(diffOfVariancesS5MinusS1),
mean(diffOfVariancesS6MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(diffOfVariancesS4MinusS2),
mean(diffOfVariancesS5MinusS2),
mean(diffOfVariancesS6MinusS2),
mean(diffOfVariancesS4MinusS3),
mean(diffOfVariancesS5MinusS3),
mean(diffOfVariancesS6MinusS3),
mean(diffOfVariancesS5MinusS4),
mean(diffOfVariancesS6MinusS4),
mean(diffOfVariancesS6MinusS5),
mean(effectSize12),
mean(effectSize13),
mean(effectSize14),
mean(effectSize15),
mean(effectSize16),
mean(effectSize23),
mean(effectSize24),
mean(effectSize25),
mean(effectSize26),
mean(effectSize34),
mean(effectSize35),
mean(effectSize36),
mean(effectSize45),
mean(effectSize46),
mean(effectSize56)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(mu4Quants[2]),
as.numeric(mu5Quants[2]),
as.numeric(mu6Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(sigma4Quants[2]),
as.numeric(sigma5Quants[2]),
as.numeric(sigma6Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu4MinusMu1Quants[2]),
as.numeric(diffOfMeansMu5MinusMu1Quants[2]),
as.numeric(diffOfMeansMu6MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu2Quants[2]),
as.numeric(diffOfMeansMu5MinusMu2Quants[2]),
as.numeric(diffOfMeansMu6MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu3Quants[2]),
as.numeric(diffOfMeansMu6MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu4Quants[2]),
as.numeric(diffOfMeansMu6MinusMu4Quants[2]),
as.numeric(diffOfMeansMu6MinusMu5Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS4MinusS1Quants[2]),
as.numeric(diffOfVariancesS5MinusS1Quants[2]),
as.numeric(diffOfVariancesS6MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS2Quants[2]),
as.numeric(diffOfVariancesS5MinusS2Quants[2]),
as.numeric(diffOfVariancesS6MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS3Quants[2]),
as.numeric(diffOfVariancesS6MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS4Quants[2]),
as.numeric(diffOfVariancesS6MinusS4Quants[2]),
as.numeric(diffOfVariancesS6MinusS5Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize14Quants[2]),
as.numeric(effectSize15Quants[2]),
as.numeric(effectSize16Quants[2]),
as.numeric(effectSize23Quants[2]),
as.numeric(effectSize24Quants[2]),
as.numeric(effectSize25Quants[2]),
as.numeric(effectSize26Quants[2]),
as.numeric(effectSize34Quants[2]),
as.numeric(effectSize35Quants[2]),
as.numeric(effectSize36Quants[2]),
as.numeric(effectSize45Quants[2]),
as.numeric(effectSize46Quants[2]),
as.numeric(effectSize56Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(mu4),
sd(mu5),
sd(mu6),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(sigma4Sq),
sd(sigma5Sq),
sd(sigma6Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu4MinusMu1),
sd(diffOfMeansMu5MinusMu1),
sd(diffOfMeansMu6MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfMeansMu4MinusMu2),
sd(diffOfMeansMu5MinusMu2),
sd(diffOfMeansMu6MinusMu2),
sd(diffOfMeansMu4MinusMu3),
sd(diffOfMeansMu5MinusMu3),
sd(diffOfMeansMu6MinusMu3),
sd(diffOfMeansMu5MinusMu4),
sd(diffOfMeansMu6MinusMu4),
sd(diffOfMeansMu6MinusMu5),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS4MinusS1),
sd(diffOfVariancesS5MinusS1),
sd(diffOfVariancesS6MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(diffOfVariancesS4MinusS2),
sd(diffOfVariancesS5MinusS2),
sd(diffOfVariancesS6MinusS2),
sd(diffOfVariancesS4MinusS3),
sd(diffOfVariancesS5MinusS3),
sd(diffOfVariancesS6MinusS3),
sd(diffOfVariancesS5MinusS4),
sd(diffOfVariancesS6MinusS4),
sd(diffOfVariancesS6MinusS5),
sd(effectSize12),
sd(effectSize13),
sd(effectSize14),
sd(effectSize15),
sd(effectSize16),
sd(effectSize23),
sd(effectSize24),
sd(effectSize25),
sd(effectSize26),
sd(effectSize34),
sd(effectSize35),
sd(effectSize36),
sd(effectSize45),
sd(effectSize46),
sd(effectSize56)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
# posterior draw dataframe
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
mu4, mu4SecC, mu4ThdC, mu4FrtC,
mu5, mu5SecC, mu5ThdC, mu5FrtC,
mu6, mu6SecC, mu6ThdC, mu6FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
sigma4Sq, sigma4SqSecC, sigma4SqThdC, sigma4SqFrtC,
sigma5Sq, sigma5SqSecC, sigma5SqThdC, sigma5SqFrtC,
sigma6Sq, sigma6SqSecC, sigma6SqThdC, sigma6SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu4MinusMu1, diffOfMeansMu4MinusMu1SecC, diffOfMeansMu4MinusMu1ThdC, diffOfMeansMu4MinusMu1FrtC,
diffOfMeansMu5MinusMu1, diffOfMeansMu5MinusMu1SecC, diffOfMeansMu5MinusMu1ThdC, diffOfMeansMu5MinusMu1FrtC,
diffOfMeansMu6MinusMu1, diffOfMeansMu6MinusMu1SecC, diffOfMeansMu6MinusMu1ThdC, diffOfMeansMu6MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfMeansMu4MinusMu2, diffOfMeansMu4MinusMu2SecC, diffOfMeansMu4MinusMu2ThdC, diffOfMeansMu4MinusMu2FrtC,
diffOfMeansMu5MinusMu2, diffOfMeansMu5MinusMu2SecC, diffOfMeansMu5MinusMu2ThdC, diffOfMeansMu5MinusMu2FrtC,
diffOfMeansMu6MinusMu2, diffOfMeansMu6MinusMu2SecC, diffOfMeansMu6MinusMu2ThdC, diffOfMeansMu6MinusMu2FrtC,
diffOfMeansMu4MinusMu3, diffOfMeansMu4MinusMu3SecC, diffOfMeansMu4MinusMu3ThdC, diffOfMeansMu4MinusMu3FrtC,
diffOfMeansMu5MinusMu3, diffOfMeansMu5MinusMu3SecC, diffOfMeansMu5MinusMu3ThdC, diffOfMeansMu5MinusMu3FrtC,
diffOfMeansMu6MinusMu3, diffOfMeansMu6MinusMu3SecC, diffOfMeansMu6MinusMu3ThdC, diffOfMeansMu6MinusMu3FrtC,
diffOfMeansMu5MinusMu4, diffOfMeansMu5MinusMu4SecC, diffOfMeansMu5MinusMu4ThdC, diffOfMeansMu5MinusMu4FrtC,
diffOfMeansMu6MinusMu4, diffOfMeansMu6MinusMu4SecC, diffOfMeansMu6MinusMu4ThdC, diffOfMeansMu6MinusMu4FrtC,
diffOfMeansMu6MinusMu5, diffOfMeansMu6MinusMu5SecC, diffOfMeansMu6MinusMu5ThdC, diffOfMeansMu6MinusMu5FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS4MinusS1, diffOfVariancesS4MinusS1SecC, diffOfVariancesS4MinusS1ThdC, diffOfVariancesS4MinusS1FrtC,
diffOfVariancesS5MinusS1, diffOfVariancesS5MinusS1SecC, diffOfVariancesS5MinusS1ThdC, diffOfVariancesS5MinusS1FrtC,
diffOfVariancesS6MinusS1, diffOfVariancesS6MinusS1SecC, diffOfVariancesS6MinusS1ThdC, diffOfVariancesS6MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
diffOfVariancesS4MinusS2, diffOfVariancesS4MinusS2SecC, diffOfVariancesS4MinusS2ThdC, diffOfVariancesS4MinusS2FrtC,
diffOfVariancesS5MinusS2, diffOfVariancesS5MinusS2SecC, diffOfVariancesS5MinusS2ThdC, diffOfVariancesS5MinusS2FrtC,
diffOfVariancesS6MinusS2, diffOfVariancesS6MinusS2SecC, diffOfVariancesS6MinusS2ThdC, diffOfVariancesS6MinusS2FrtC,
diffOfVariancesS4MinusS3, diffOfVariancesS4MinusS3SecC, diffOfVariancesS4MinusS3ThdC, diffOfVariancesS4MinusS3FrtC,
diffOfVariancesS5MinusS3, diffOfVariancesS5MinusS3SecC, diffOfVariancesS5MinusS3ThdC, diffOfVariancesS5MinusS3FrtC,
diffOfVariancesS6MinusS3, diffOfVariancesS6MinusS3SecC, diffOfVariancesS6MinusS3ThdC, diffOfVariancesS6MinusS3FrtC,
diffOfVariancesS5MinusS4, diffOfVariancesS5MinusS4SecC, diffOfVariancesS5MinusS4ThdC, diffOfVariancesS5MinusS4FrtC,
diffOfVariancesS6MinusS4, diffOfVariancesS6MinusS4SecC, diffOfVariancesS6MinusS4ThdC, diffOfVariancesS6MinusS4FrtC,
diffOfVariancesS6MinusS5, diffOfVariancesS6MinusS5SecC, diffOfVariancesS6MinusS5ThdC, diffOfVariancesS6MinusS5FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize14, effectSize14SecC, effectSize14ThdC, effectSize14FrtC,
effectSize15, effectSize15SecC, effectSize15ThdC, effectSize15FrtC,
effectSize16, effectSize16SecC, effectSize16ThdC, effectSize16FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC,
effectSize24, effectSize24SecC, effectSize24ThdC, effectSize24FrtC,
effectSize25, effectSize25SecC, effectSize25ThdC, effectSize25FrtC,
effectSize26, effectSize26SecC, effectSize26ThdC, effectSize26FrtC,
effectSize34, effectSize34SecC, effectSize34ThdC, effectSize34FrtC,
effectSize35, effectSize35SecC, effectSize35ThdC, effectSize35FrtC,
effectSize36, effectSize36SecC, effectSize36ThdC, effectSize36FrtC,
effectSize45, effectSize45SecC, effectSize45ThdC, effectSize45FrtC,
effectSize46, effectSize46SecC, effectSize46ThdC, effectSize46FrtC,
effectSize56, effectSize56SecC, effectSize56ThdC, effectSize56FrtC)
#pretty table
print(knitr::kable(out))
# return dataframe
return(df)
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && length(fifth)>1 && is.null(sixth)) {
Parameter<-c("mu1",
"mu2",
"mu3",
"mu4",
"mu5",
"sigma1",
"sigma2",
"sigma3",
"sigma4",
"sigma5",
"mu2-mu1",
"mu3-mu1",
"mu4-mu1",
"mu5-mu1",
"mu3-mu2",
"mu4-mu2",
"mu5-mu2",
"mu4-mu3",
"mu5-mu3",
"mu5-mu4",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma4-sigma1",
"sigma5-sigma1",
"sigma3-sigma2",
"sigma4-sigma2",
"sigma5-sigma2",
"sigma4-sigma3",
"sigma5-sigma3",
"sigma5-sigma4",
"delta12",
"delta13",
"delta14",
"delta15",
"delta23",
"delta24",
"delta25",
"delta34",
"delta35",
"delta45")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(mu4Quants[1]),
as.numeric(mu5Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(sigma4Quants[1]),
as.numeric(sigma5Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu4MinusMu1Quants[1]),
as.numeric(diffOfMeansMu5MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu2Quants[1]),
as.numeric(diffOfMeansMu5MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu3Quants[1]),
as.numeric(diffOfMeansMu5MinusMu4Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS4MinusS1Quants[1]),
as.numeric(diffOfVariancesS5MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS2Quants[1]),
as.numeric(diffOfVariancesS5MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS3Quants[1]),
as.numeric(diffOfVariancesS5MinusS4Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize14Quants[1]),
as.numeric(effectSize15Quants[1]),
as.numeric(effectSize23Quants[1]),
as.numeric(effectSize24Quants[1]),
as.numeric(effectSize25Quants[1]),
as.numeric(effectSize34Quants[1]),
as.numeric(effectSize35Quants[1]),
as.numeric(effectSize45Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(mu4),
mean(mu5),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(sigma4Sq),
mean(sigma5Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu4MinusMu1),
mean(diffOfMeansMu5MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfMeansMu4MinusMu2),
mean(diffOfMeansMu5MinusMu2),
mean(diffOfMeansMu4MinusMu3),
mean(diffOfMeansMu5MinusMu3),
mean(diffOfMeansMu5MinusMu4),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS4MinusS1),
mean(diffOfVariancesS5MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(diffOfVariancesS4MinusS2),
mean(diffOfVariancesS5MinusS2),
mean(diffOfVariancesS4MinusS3),
mean(diffOfVariancesS5MinusS3),
mean(diffOfVariancesS5MinusS4),
mean(effectSize12),
mean(effectSize13),
mean(effectSize14),
mean(effectSize15),
mean(effectSize23),
mean(effectSize24),
mean(effectSize25),
mean(effectSize34),
mean(effectSize35),
mean(effectSize45)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(mu4Quants[2]),
as.numeric(mu5Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(sigma4Quants[2]),
as.numeric(sigma5Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu4MinusMu1Quants[2]),
as.numeric(diffOfMeansMu5MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu2Quants[2]),
as.numeric(diffOfMeansMu5MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu3Quants[2]),
as.numeric(diffOfMeansMu5MinusMu4Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS4MinusS1Quants[2]),
as.numeric(diffOfVariancesS5MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS2Quants[2]),
as.numeric(diffOfVariancesS5MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS3Quants[2]),
as.numeric(diffOfVariancesS5MinusS4Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize14Quants[2]),
as.numeric(effectSize15Quants[2]),
as.numeric(effectSize23Quants[2]),
as.numeric(effectSize24Quants[2]),
as.numeric(effectSize25Quants[2]),
as.numeric(effectSize34Quants[2]),
as.numeric(effectSize35Quants[2]),
as.numeric(effectSize45Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(mu4),
sd(mu5),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(sigma4Sq),
sd(sigma5Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu4MinusMu1),
sd(diffOfMeansMu5MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfMeansMu4MinusMu2),
sd(diffOfMeansMu5MinusMu2),
sd(diffOfMeansMu4MinusMu3),
sd(diffOfMeansMu5MinusMu3),
sd(diffOfMeansMu5MinusMu4),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS4MinusS1),
sd(diffOfVariancesS5MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(diffOfVariancesS4MinusS2),
sd(diffOfVariancesS5MinusS2),
sd(diffOfVariancesS4MinusS3),
sd(diffOfVariancesS5MinusS3),
sd(diffOfVariancesS5MinusS4),
sd(effectSize12),
sd(effectSize13),
sd(effectSize14),
sd(effectSize15),
sd(effectSize23),
sd(effectSize24),
sd(effectSize25),
sd(effectSize34),
sd(effectSize35),
sd(effectSize45)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
# posterior draw dataframe
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
mu4, mu4SecC, mu4ThdC, mu4FrtC,
mu5, mu5SecC, mu5ThdC, mu5FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
sigma4Sq, sigma4SqSecC, sigma4SqThdC, sigma4SqFrtC,
sigma5Sq, sigma5SqSecC, sigma5SqThdC, sigma5SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu4MinusMu1, diffOfMeansMu4MinusMu1SecC, diffOfMeansMu4MinusMu1ThdC, diffOfMeansMu4MinusMu1FrtC,
diffOfMeansMu5MinusMu1, diffOfMeansMu5MinusMu1SecC, diffOfMeansMu5MinusMu1ThdC, diffOfMeansMu5MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfMeansMu4MinusMu2, diffOfMeansMu4MinusMu2SecC, diffOfMeansMu4MinusMu2ThdC, diffOfMeansMu4MinusMu2FrtC,
diffOfMeansMu5MinusMu2, diffOfMeansMu5MinusMu2SecC, diffOfMeansMu5MinusMu2ThdC, diffOfMeansMu5MinusMu2FrtC,
diffOfMeansMu4MinusMu3, diffOfMeansMu4MinusMu3SecC, diffOfMeansMu4MinusMu3ThdC, diffOfMeansMu4MinusMu3FrtC,
diffOfMeansMu5MinusMu3, diffOfMeansMu5MinusMu3SecC, diffOfMeansMu5MinusMu3ThdC, diffOfMeansMu5MinusMu3FrtC,
diffOfMeansMu5MinusMu4, diffOfMeansMu5MinusMu4SecC, diffOfMeansMu5MinusMu4ThdC, diffOfMeansMu5MinusMu4FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS4MinusS1, diffOfVariancesS4MinusS1SecC, diffOfVariancesS4MinusS1ThdC, diffOfVariancesS4MinusS1FrtC,
diffOfVariancesS5MinusS1, diffOfVariancesS5MinusS1SecC, diffOfVariancesS5MinusS1ThdC, diffOfVariancesS5MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
diffOfVariancesS4MinusS2, diffOfVariancesS4MinusS2SecC, diffOfVariancesS4MinusS2ThdC, diffOfVariancesS4MinusS2FrtC,
diffOfVariancesS5MinusS2, diffOfVariancesS5MinusS2SecC, diffOfVariancesS5MinusS2ThdC, diffOfVariancesS5MinusS2FrtC,
diffOfVariancesS4MinusS3, diffOfVariancesS4MinusS3SecC, diffOfVariancesS4MinusS3ThdC, diffOfVariancesS4MinusS3FrtC,
diffOfVariancesS5MinusS3, diffOfVariancesS5MinusS3SecC, diffOfVariancesS5MinusS3ThdC, diffOfVariancesS5MinusS3FrtC,
diffOfVariancesS5MinusS4, diffOfVariancesS5MinusS4SecC, diffOfVariancesS5MinusS4ThdC, diffOfVariancesS5MinusS4FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize14, effectSize14SecC, effectSize14ThdC, effectSize14FrtC,
effectSize15, effectSize15SecC, effectSize15ThdC, effectSize15FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC,
effectSize24, effectSize24SecC, effectSize24ThdC, effectSize24FrtC,
effectSize25, effectSize25SecC, effectSize25ThdC, effectSize25FrtC,
effectSize34, effectSize34SecC, effectSize34ThdC, effectSize34FrtC,
effectSize35, effectSize35SecC, effectSize35ThdC, effectSize35FrtC,
effectSize45, effectSize45SecC, effectSize45ThdC, effectSize45FrtC)
#pretty table
print(knitr::kable(out))
# return posterior draw dataframe
return(df)
} else if(length(first)>1 && length(second)>1 && length(third)>1 && length(fourth)>1 && is.null(fifth) && is.null(sixth)) {
Parameter<-c("mu1",
"mu2",
"mu3",
"mu4",
"sigma1",
"sigma2",
"sigma3",
"sigma4",
"mu2-mu1",
"mu3-mu1",
"mu4-mu1",
"mu3-mu2",
"mu4-mu2",
"mu4-mu3",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma4-sigma1",
"sigma3-sigma2",
"sigma4-sigma2",
"sigma4-sigma3",
"delta12",
"delta13",
"delta14",
"delta23",
"delta24",
"delta34")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(mu4Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(sigma4Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu4MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu2Quants[1]),
as.numeric(diffOfMeansMu4MinusMu3Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS4MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS2Quants[1]),
as.numeric(diffOfVariancesS4MinusS3Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize14Quants[1]),
as.numeric(effectSize23Quants[1]),
as.numeric(effectSize24Quants[1]),
as.numeric(effectSize34Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(mu4),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(sigma4Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu4MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfMeansMu4MinusMu2),
mean(diffOfMeansMu4MinusMu3),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS4MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(diffOfVariancesS4MinusS2),
mean(diffOfVariancesS4MinusS3),
mean(effectSize12),
mean(effectSize13),
mean(effectSize14),
mean(effectSize23),
mean(effectSize24),
mean(effectSize34)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(mu4Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(sigma4Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu4MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu2Quants[2]),
as.numeric(diffOfMeansMu4MinusMu3Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS4MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS2Quants[2]),
as.numeric(diffOfVariancesS4MinusS3Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize14Quants[2]),
as.numeric(effectSize23Quants[2]),
as.numeric(effectSize24Quants[2]),
as.numeric(effectSize34Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(mu4),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(sigma4Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu4MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfMeansMu4MinusMu2),
sd(diffOfMeansMu4MinusMu3),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS4MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(diffOfVariancesS4MinusS2),
sd(diffOfVariancesS4MinusS3),
sd(effectSize12),
sd(effectSize13),
sd(effectSize14),
sd(effectSize23),
sd(effectSize24),
sd(effectSize34)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
#pretty table
print(knitr::kable(out))
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
mu4, mu4SecC, mu4ThdC, mu4FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
sigma4Sq, sigma4SqSecC, sigma4SqThdC, sigma4SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu4MinusMu1, diffOfMeansMu4MinusMu1SecC, diffOfMeansMu4MinusMu1ThdC, diffOfMeansMu4MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfMeansMu4MinusMu2, diffOfMeansMu4MinusMu2SecC, diffOfMeansMu4MinusMu2ThdC, diffOfMeansMu4MinusMu2FrtC,
diffOfMeansMu4MinusMu3, diffOfMeansMu4MinusMu3SecC, diffOfMeansMu4MinusMu3ThdC, diffOfMeansMu4MinusMu3FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS4MinusS1, diffOfVariancesS4MinusS1SecC, diffOfVariancesS4MinusS1ThdC, diffOfVariancesS4MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
diffOfVariancesS4MinusS2, diffOfVariancesS4MinusS2SecC, diffOfVariancesS4MinusS2ThdC, diffOfVariancesS4MinusS2FrtC,
diffOfVariancesS4MinusS3, diffOfVariancesS4MinusS3SecC, diffOfVariancesS4MinusS3ThdC, diffOfVariancesS4MinusS3FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize14, effectSize14SecC, effectSize14ThdC, effectSize14FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC,
effectSize24, effectSize24SecC, effectSize24ThdC, effectSize24FrtC,
effectSize34, effectSize34SecC, effectSize34ThdC, effectSize34FrtC)
return(df)
} else if(length(first)>1 && length(second)>1 && length(third)>1 && is.null(fourth) && is.null(fifth) && is.null(sixth)) {
Parameter<-c("mu1",
"mu2",
"mu3",
"sigma1",
"sigma2",
"sigma3",
"mu2-mu1",
"mu3-mu1",
"mu3-mu2",
"sigma2-sigma1",
"sigma3-sigma1",
"sigma3-sigma2",
"delta12",
"delta13",
"delta23")
LQ<-round(c(as.numeric(mu1Quants[1]),
as.numeric(mu2Quants[1]),
as.numeric(mu3Quants[1]),
as.numeric(sigma1Quants[1]),
as.numeric(sigma2Quants[1]),
as.numeric(sigma3Quants[1]),
as.numeric(diffOfMeansMu2MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu1Quants[1]),
as.numeric(diffOfMeansMu3MinusMu2Quants[1]),
as.numeric(diffOfVariancesS2MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS1Quants[1]),
as.numeric(diffOfVariancesS3MinusS2Quants[1]),
as.numeric(effectSize12Quants[1]),
as.numeric(effectSize13Quants[1]),
as.numeric(effectSize23Quants[1])),2)
Mean<-round(c(mean(mu1),
mean(mu2),
mean(mu3),
mean(sigma1Sq),
mean(sigma2Sq),
mean(sigma3Sq),
mean(diffOfMeansMu2MinusMu1),
mean(diffOfMeansMu3MinusMu1),
mean(diffOfMeansMu3MinusMu2),
mean(diffOfVariancesS2MinusS1),
mean(diffOfVariancesS3MinusS1),
mean(diffOfVariancesS3MinusS2),
mean(effectSize12),
mean(effectSize13),
mean(effectSize23)),2)
UQ<-round(c(as.numeric(mu1Quants[2]),
as.numeric(mu2Quants[2]),
as.numeric(mu3Quants[2]),
as.numeric(sigma1Quants[2]),
as.numeric(sigma2Quants[2]),
as.numeric(sigma3Quants[2]),
as.numeric(diffOfMeansMu2MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu1Quants[2]),
as.numeric(diffOfMeansMu3MinusMu2Quants[2]),
as.numeric(diffOfVariancesS2MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS1Quants[2]),
as.numeric(diffOfVariancesS3MinusS2Quants[2]),
as.numeric(effectSize12Quants[2]),
as.numeric(effectSize13Quants[2]),
as.numeric(effectSize23Quants[2])),2)
Std.Err<-round(c(sd(mu1),
sd(mu2),
sd(mu3),
sd(sigma1Sq),
sd(sigma2Sq),
sd(sigma3Sq),
sd(diffOfMeansMu2MinusMu1),
sd(diffOfMeansMu3MinusMu1),
sd(diffOfMeansMu3MinusMu2),
sd(diffOfVariancesS2MinusS1),
sd(diffOfVariancesS3MinusS1),
sd(diffOfVariancesS3MinusS2),
sd(effectSize12),
sd(effectSize13),
sd(effectSize23)),2)
#dataframe output
out <- cbind(Parameter,LQ,Mean,UQ,Std.Err)
#pretty table
print(knitr::kable(out))
df=data.frame(mu1, mu1SecC, mu1ThdC, mu1FrtC,
mu2, mu2SecC, mu2ThdC, mu2FrtC,
mu3, mu3SecC, mu3ThdC, mu3FrtC,
sigma1Sq, sigma1SqSecC, sigma1SqThdC, sigma1SqFrtC,
sigma2Sq, sigma2SqSecC, sigma2SqThdC, sigma2SqFrtC,
sigma3Sq, sigma3SqSecC, sigma3SqThdC, sigma3SqFrtC,
diffOfMeansMu2MinusMu1, diffOfMeansMu2MinusMu1SecC, diffOfMeansMu2MinusMu1ThdC, diffOfMeansMu2MinusMu1FrtC,
diffOfMeansMu3MinusMu1, diffOfMeansMu3MinusMu1SecC, diffOfMeansMu3MinusMu1ThdC, diffOfMeansMu3MinusMu1FrtC,
diffOfMeansMu3MinusMu2, diffOfMeansMu3MinusMu2SecC, diffOfMeansMu3MinusMu2ThdC, diffOfMeansMu3MinusMu2FrtC,
diffOfVariancesS2MinusS1, diffOfVariancesS2MinusS1SecC, diffOfVariancesS2MinusS1ThdC, diffOfVariancesS2MinusS1FrtC,
diffOfVariancesS3MinusS1, diffOfVariancesS3MinusS1SecC, diffOfVariancesS3MinusS1ThdC, diffOfVariancesS3MinusS1FrtC,
diffOfVariancesS3MinusS2, diffOfVariancesS3MinusS2SecC, diffOfVariancesS3MinusS2ThdC, diffOfVariancesS3MinusS2FrtC,
effectSize12, effectSize12SecC, effectSize12ThdC, effectSize12FrtC,
effectSize13, effectSize13SecC, effectSize13ThdC, effectSize13FrtC,
effectSize23, effectSize23SecC, effectSize23ThdC, effectSize23FrtC)
return(df)
}
}
anovaplot = function(dataframe, type="rope", sd="sd", ci=0.95){
# load parameter chains from dataframe
mu1=dataframe$mu1; mu1SecC=dataframe$mu1SecC; mu1ThdC=dataframe$mu1ThdC; mu1FrtC=dataframe$mu1FrtC;
mu2=dataframe$mu2; mu2SecC=dataframe$mu2SecC; mu2ThdC=dataframe$mu2ThdC; mu2FrtC=dataframe$mu2FrtC;
mu3=dataframe$mu3; mu3SecC=dataframe$mu3SecC; mu3ThdC=dataframe$mu3ThdC; mu3FrtC=dataframe$mu3FrtC;
if(!is.null(dataframe$mu4)){
mu4=dataframe$mu4; mu4SecC=dataframe$mu4SecC; mu4ThdC=dataframe$mu4ThdC; mu4FrtC=dataframe$mu4FrtC;
}
if(!is.null(dataframe$mu5)){
mu5=dataframe$mu5; mu5SecC=dataframe$mu5SecC; mu5ThdC=dataframe$mu5ThdC; mu5FrtC=dataframe$mu5FrtC;
}
if(!is.null(dataframe$mu6)){
mu6=dataframe$mu6; mu6SecC=dataframe$mu6SecC; mu6ThdC=dataframe$mu6ThdC; mu6FrtC=dataframe$mu6FrtC;
}
sigma1Sq=dataframe$sigma1Sq; sigma1SqSecC=dataframe$sigma1SqSecC; sigma1SqThdC=dataframe$sigma1SqThdC; sigma1SqFrtC=dataframe$sigma1SqFrtC;
sigma2Sq=dataframe$sigma2Sq; sigma2SqSecC=dataframe$sigma2SqSecC; sigma2SqThdC=dataframe$sigma2SqThdC; sigma2SqFrtC=dataframe$sigma2SqFrtC;
sigma3Sq=dataframe$sigma3Sq; sigma3SqSecC=dataframe$sigma3SqSecC; sigma3SqThdC=dataframe$sigma3SqThdC; sigma3SqFrtC=dataframe$sigma3SqFrtC;
if(!is.null(dataframe$sigma4Sq)){
sigma4Sq=dataframe$sigma4Sq; sigma4SqSecC=dataframe$sigma4SqSecC; sigma4SqThdC=dataframe$sigma4SqThdC; sigma4SqFrtC=dataframe$sigma4SqFrtC;
}
if(!is.null(dataframe$sigma5Sq)){
sigma5Sq=dataframe$sigma5Sq; sigma5SqSecC=dataframe$sigma5SqSecC; sigma5SqThdC=dataframe$sigma5SqThdC; sigma5SqFrtC=dataframe$sigma5SqFrtC;
}
if(!is.null(dataframe$sigma6Sq)){
sigma6Sq=dataframe$sigma6Sq; sigma6SqSecC=dataframe$sigma6SqSecC; sigma6SqThdC=dataframe$sigma6SqThdC; sigma6SqFrtC=dataframe$sigma6SqFrtC;
}
diffOfMeansMu2MinusMu1=dataframe$diffOfMeansMu2MinusMu1; diffOfMeansMu2MinusMu1SecC=dataframe$diffOfMeansMu2MinusMu1SecC; diffOfMeansMu2MinusMu1ThdC=dataframe$diffOfMeansMu2MinusMu1ThdC; diffOfMeansMu2MinusMu1FrtC=dataframe$diffOfMeansMu2MinusMu1FrtC;
diffOfMeansMu3MinusMu1=dataframe$diffOfMeansMu3MinusMu1; diffOfMeansMu3MinusMu1SecC=dataframe$diffOfMeansMu3MinusMu1SecC; diffOfMeansMu3MinusMu1ThdC=dataframe$diffOfMeansMu3MinusMu1ThdC; diffOfMeansMu3MinusMu1FrtC=dataframe$diffOfMeansMu3MinusMu1FrtC;
diffOfMeansMu3MinusMu2=dataframe$diffOfMeansMu3MinusMu2;diffOfMeansMu3MinusMu2SecC=dataframe$diffOfMeansMu3MinusMu2SecC; diffOfMeansMu3MinusMu2ThdC=dataframe$diffOfMeansMu3MinusMu2ThdC; diffOfMeansMu3MinusMu2FrtC=dataframe$diffOfMeansMu3MinusMu2FrtC;
if(!is.null(dataframe$sigma4Sq)){
diffOfMeansMu4MinusMu1=dataframe$diffOfMeansMu4MinusMu1; diffOfMeansMu4MinusMu1SecC=dataframe$diffOfMeansMu4MinusMu1SecC; diffOfMeansMu4MinusMu1ThdC=dataframe$diffOfMeansMu4MinusMu1ThdC; diffOfMeansMu4MinusMu1FrtC=dataframe$diffOfMeansMu4MinusMu1FrtC;
diffOfMeansMu4MinusMu2=dataframe$diffOfMeansMu4MinusMu2; diffOfMeansMu4MinusMu2SecC=dataframe$diffOfMeansMu4MinusMu2SecC; diffOfMeansMu4MinusMu2ThdC=dataframe$diffOfMeansMu4MinusMu2ThdC; diffOfMeansMu4MinusMu2FrtC=dataframe$diffOfMeansMu4MinusMu2FrtC;
diffOfMeansMu4MinusMu3=dataframe$diffOfMeansMu4MinusMu3; diffOfMeansMu4MinusMu3SecC=dataframe$diffOfMeansMu4MinusMu3SecC; diffOfMeansMu4MinusMu3ThdC=dataframe$diffOfMeansMu4MinusMu3ThdC; diffOfMeansMu4MinusMu3FrtC=dataframe$diffOfMeansMu4MinusMu3FrtC;
}
if(!is.null(dataframe$sigma5Sq)){
diffOfMeansMu5MinusMu1=dataframe$diffOfMeansMu5MinusMu1; diffOfMeansMu5MinusMu1SecC=dataframe$diffOfMeansMu5MinusMu1SecC; diffOfMeansMu5MinusMu1ThdC=dataframe$diffOfMeansMu5MinusMu1ThdC; diffOfMeansMu5MinusMu1FrtC=dataframe$diffOfMeansMu5MinusMu1FrtC;
diffOfMeansMu5MinusMu2=dataframe$diffOfMeansMu5MinusMu2; diffOfMeansMu5MinusMu2SecC=dataframe$diffOfMeansMu5MinusMu2SecC; diffOfMeansMu5MinusMu2ThdC=dataframe$diffOfMeansMu5MinusMu2ThdC; diffOfMeansMu5MinusMu2FrtC=dataframe$diffOfMeansMu5MinusMu2FrtC;
diffOfMeansMu5MinusMu3=dataframe$diffOfMeansMu5MinusMu3; diffOfMeansMu5MinusMu3SecC=dataframe$diffOfMeansMu5MinusMu3SecC; diffOfMeansMu5MinusMu3ThdC=dataframe$diffOfMeansMu5MinusMu3ThdC; diffOfMeansMu5MinusMu3FrtC=dataframe$diffOfMeansMu5MinusMu3FrtC;
diffOfMeansMu5MinusMu4=dataframe$diffOfMeansMu5MinusMu4; diffOfMeansMu5MinusMu4SecC=dataframe$diffOfMeansMu5MinusMu4SecC; diffOfMeansMu5MinusMu4ThdC=dataframe$diffOfMeansMu5MinusMu4ThdC; diffOfMeansMu5MinusMu4FrtC=dataframe$diffOfMeansMu5MinusMu4FrtC;
}
if(!is.null(dataframe$sigma6Sq)){
diffOfMeansMu6MinusMu1=dataframe$diffOfMeansMu6MinusMu1; diffOfMeansMu6MinusMu1SecC=dataframe$diffOfMeansMu6MinusMu1SecC; diffOfMeansMu6MinusMu1ThdC=dataframe$diffOfMeansMu6MinusMu1ThdC; diffOfMeansMu6MinusMu1FrtC=dataframe$diffOfMeansMu6MinusMu1FrtC;
diffOfMeansMu6MinusMu2=dataframe$diffOfMeansMu6MinusMu2; diffOfMeansMu6MinusMu2SecC=dataframe$diffOfMeansMu6MinusMu2SecC; diffOfMeansMu6MinusMu2ThdC=dataframe$diffOfMeansMu6MinusMu2ThdC; diffOfMeansMu6MinusMu2FrtC=dataframe$diffOfMeansMu6MinusMu2FrtC;
diffOfMeansMu6MinusMu3=dataframe$diffOfMeansMu6MinusMu3; diffOfMeansMu6MinusMu3SecC=dataframe$diffOfMeansMu6MinusMu3SecC; diffOfMeansMu6MinusMu3ThdC=dataframe$diffOfMeansMu6MinusMu3ThdC; diffOfMeansMu6MinusMu3FrtC=dataframe$diffOfMeansMu6MinusMu3FrtC;
diffOfMeansMu6MinusMu4=dataframe$diffOfMeansMu6MinusMu4; diffOfMeansMu6MinusMu4SecC=dataframe$diffOfMeansMu6MinusMu4SecC; diffOfMeansMu6MinusMu4ThdC=dataframe$diffOfMeansMu6MinusMu4ThdC; diffOfMeansMu6MinusMu4FrtC=dataframe$diffOfMeansMu6MinusMu4FrtC;
diffOfMeansMu6MinusMu5=dataframe$diffOfMeansMu6MinusMu5; diffOfMeansMu6MinusMu5SecC=dataframe$diffOfMeansMu6MinusMu5SecC; diffOfMeansMu6MinusMu5ThdC=dataframe$diffOfMeansMu6MinusMu5ThdC; diffOfMeansMu6MinusMu5FrtC=dataframe$diffOfMeansMu6MinusMu5FrtC;
}
diffOfVariancesS2MinusS1=dataframe$diffOfVariancesS2MinusS1; diffOfVariancesS2MinusS1SecC=dataframe$diffOfVariancesS2MinusS1SecC; diffOfVariancesS2MinusS1ThdC=dataframe$diffOfVariancesS2MinusS1ThdC; diffOfVariancesS2MinusS1FrtC=dataframe$diffOfVariancesS2MinusS1FrtC;
diffOfVariancesS3MinusS1=dataframe$diffOfVariancesS3MinusS1; diffOfVariancesS3MinusS1SecC=dataframe$diffOfVariancesS3MinusS1SecC; diffOfVariancesS3MinusS1ThdC=dataframe$diffOfVariancesS3MinusS1ThdC; diffOfVariancesS3MinusS1FrtC=dataframe$diffOfVariancesS3MinusS1FrtC;
diffOfVariancesS3MinusS2=dataframe$diffOfVariancesS3MinusS2;diffOfVariancesS3MinusS2SecC=dataframe$diffOfVariancesS3MinusS2SecC; diffOfVariancesS3MinusS2ThdC=dataframe$diffOfVariancesS3MinusS2ThdC; diffOfVariancesS3MinusS2FrtC=dataframe$diffOfVariancesS3MinusS2FrtC;
if(!is.null(dataframe$sigma4Sq)){
diffOfVariancesS4MinusS1=dataframe$diffOfVariancesS4MinusS1; diffOfVariancesS4MinusS1SecC=dataframe$diffOfVariancesS4MinusS1SecC; diffOfVariancesS4MinusS1ThdC=dataframe$diffOfVariancesS4MinusS1ThdC; diffOfVariancesS4MinusS1FrtC=dataframe$diffOfVariancesS4MinusS1FrtC;
diffOfVariancesS4MinusS2=dataframe$diffOfVariancesS4MinusS2; diffOfVariancesS4MinusS2SecC=dataframe$diffOfVariancesS4MinusS2SecC; diffOfVariancesS4MinusS2ThdC=dataframe$diffOfVariancesS4MinusS2ThdC; diffOfVariancesS4MinusS2FrtC=dataframe$diffOfVariancesS4MinusS2FrtC;
diffOfVariancesS4MinusS3=dataframe$diffOfVariancesS4MinusS3; diffOfVariancesS4MinusS3SecC=dataframe$diffOfVariancesS4MinusS3SecC; diffOfVariancesS4MinusS3ThdC=dataframe$diffOfVariancesS4MinusS3ThdC; diffOfVariancesS4MinusS3FrtC=dataframe$diffOfVariancesS4MinusS3FrtC;
}
if(!is.null(dataframe$sigma5Sq)){
diffOfVariancesS5MinusS1=dataframe$diffOfVariancesS5MinusS1; diffOfVariancesS5MinusS1SecC=dataframe$diffOfVariancesS5MinusS1SecC; diffOfVariancesS5MinusS1ThdC=dataframe$diffOfVariancesS5MinusS1ThdC; diffOfVariancesS5MinusS1FrtC=dataframe$diffOfVariancesS5MinusS1FrtC;
diffOfVariancesS5MinusS2=dataframe$diffOfVariancesS5MinusS2; diffOfVariancesS5MinusS2SecC=dataframe$diffOfVariancesS5MinusS2SecC; diffOfVariancesS5MinusS2ThdC=dataframe$diffOfVariancesS5MinusS2ThdC; diffOfVariancesS5MinusS2FrtC=dataframe$diffOfVariancesS5MinusS2FrtC;
diffOfVariancesS5MinusS3=dataframe$diffOfVariancesS5MinusS3; diffOfVariancesS5MinusS3SecC=dataframe$diffOfVariancesS5MinusS3SecC; diffOfVariancesS5MinusS3ThdC=dataframe$diffOfVariancesS5MinusS3ThdC; diffOfVariancesS5MinusS3FrtC=dataframe$diffOfVariancesS5MinusS3FrtC;
diffOfVariancesS5MinusS4=dataframe$diffOfVariancesS5MinusS4; diffOfVariancesS5MinusS4SecC=dataframe$diffOfVariancesS5MinusS4SecC; diffOfVariancesS5MinusS4ThdC=dataframe$diffOfVariancesS5MinusS4ThdC; diffOfVariancesS5MinusS4FrtC=dataframe$diffOfVariancesS5MinusS4FrtC;
}
if(!is.null(dataframe$sigma6Sq)){
diffOfVariancesS6MinusS1=dataframe$diffOfVariancesS6MinusS1; diffOfVariancesS6MinusS1SecC=dataframe$diffOfVariancesS6MinusS1SecC; diffOfVariancesS6MinusS1ThdC=dataframe$diffOfVariancesS6MinusS1ThdC; diffOfVariancesS6MinusS1FrtC=dataframe$diffOfVariancesS6MinusS1FrtC;
diffOfVariancesS6MinusS2=dataframe$diffOfVariancesS6MinusS2; diffOfVariancesS6MinusS2SecC=dataframe$diffOfVariancesS6MinusS2SecC; diffOfVariancesS6MinusS2ThdC=dataframe$diffOfVariancesS6MinusS2ThdC; diffOfVariancesS6MinusS2FrtC=dataframe$diffOfVariancesS6MinusS2FrtC;
diffOfVariancesS6MinusS3=dataframe$diffOfVariancesS6MinusS3; diffOfVariancesS6MinusS3SecC=dataframe$diffOfVariancesS6MinusS3SecC; diffOfVariancesS6MinusS3ThdC=dataframe$diffOfVariancesS6MinusS3ThdC; diffOfVariancesS6MinusS3FrtC=dataframe$diffOfVariancesS6MinusS3FrtC;
diffOfVariancesS6MinusS4=dataframe$diffOfVariancesS6MinusS4; diffOfVariancesS6MinusS4SecC=dataframe$diffOfVariancesS6MinusS4SecC; diffOfVariancesS6MinusS4ThdC=dataframe$diffOfVariancesS6MinusS4ThdC; diffOfVariancesS6MinusS4FrtC=dataframe$diffOfVariancesS6MinusS4FrtC;
diffOfVariancesS6MinusS5=dataframe$diffOfVariancesS6MinusS5; diffOfVariancesS6MinusS5SecC=dataframe$diffOfVariancesS6MinusS5SecC; diffOfVariancesS6MinusS5ThdC=dataframe$diffOfVariancesS6MinusS5ThdC; diffOfVariancesS6MinusS5FrtC=dataframe$diffOfVariancesS6MinusS5FrtC;
}
effectSize12=dataframe$effectSize12; effectSize12SecC=dataframe$effectSize12SecC; effectSize12ThdC=dataframe$effectSize12ThdC; effectSize12FrtC=dataframe$effectSize12FrtC;
effectSize13=dataframe$effectSize13; effectSize13SecC=dataframe$effectSize13SecC; effectSize13ThdC=dataframe$effectSize13ThdC; effectSize13FrtC=dataframe$effectSize13FrtC;
effectSize23=dataframe$effectSize23; effectSize23SecC=dataframe$effectSize23SecC; effectSize23ThdC=dataframe$effectSize23ThdC; effectSize23FrtC=dataframe$effectSize23FrtC;
if(!is.null(dataframe$sigma4Sq)){
effectSize14=dataframe$effectSize14; effectSize14SecC=dataframe$effectSize14SecC; effectSize14ThdC=dataframe$effectSize14ThdC; effectSize14FrtC=dataframe$effectSize14FrtC;
effectSize24=dataframe$effectSize24; effectSize24SecC=dataframe$effectSize24SecC; effectSize24ThdC=dataframe$effectSize24ThdC; effectSize24FrtC=dataframe$effectSize24FrtC;
effectSize34=dataframe$effectSize34; effectSize34SecC=dataframe$effectSize34SecC; effectSize34ThdC=dataframe$effectSize34ThdC; effectSize34FrtC=dataframe$effectSize34FrtC;
}
if(!is.null(dataframe$sigma5Sq)){
effectSize15=dataframe$effectSize15; effectSize15SecC=dataframe$effectSize15SecC; effectSize15ThdC=dataframe$effectSize15ThdC; effectSize15FrtC=dataframe$effectSize15FrtC;
effectSize25=dataframe$effectSize25; effectSize25SecC=dataframe$effectSize25SecC; effectSize25ThdC=dataframe$effectSize25ThdC; effectSize25FrtC=dataframe$effectSize25FrtC;
effectSize35=dataframe$effectSize35; effectSize35SecC=dataframe$effectSize35SecC; effectSize35ThdC=dataframe$effectSize35ThdC; effectSize35FrtC=dataframe$effectSize35FrtC;
effectSize45=dataframe$effectSize45; effectSize45SecC=dataframe$effectSize45SecC; effectSize45ThdC=dataframe$effectSize45ThdC; effectSize45FrtC=dataframe$effectSize45FrtC;
}
if(!is.null(dataframe$sigma6Sq)){
effectSize16=dataframe$effectSize16; effectSize16SecC=dataframe$effectSize16SecC; effectSize16ThdC=dataframe$effectSize16ThdC; effectSize16FrtC=dataframe$effectSize16FrtC;
effectSize26=dataframe$effectSize26; effectSize26SecC=dataframe$effectSize26SecC; effectSize26ThdC=dataframe$effectSize26ThdC; effectSize26FrtC=dataframe$effectSize26FrtC;
effectSize36=dataframe$effectSize36; effectSize36SecC=dataframe$effectSize36SecC; effectSize36ThdC=dataframe$effectSize36ThdC; effectSize36FrtC=dataframe$effectSize36FrtC;
effectSize46=dataframe$effectSize46; effectSize46SecC=dataframe$effectSize46SecC; effectSize46ThdC=dataframe$effectSize46ThdC; effectSize46FrtC=dataframe$effectSize46FrtC;
effectSize56=dataframe$effectSize56; effectSize56SecC=dataframe$effectSize56SecC; effectSize56ThdC=dataframe$effectSize56ThdC; effectSize56FrtC=dataframe$effectSize56FrtC;
}
# posterior plots for each parameter
if(type=="pars"){
# MU1
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu1,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[1]))
plot(mu1,ty="l",col="cornflowerblue",xlab=expression(mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=mu1)
mcmcObj2=coda::mcmc(data=mu1SecC)
mcmcObj3=coda::mcmc(data=mu1ThdC)
mcmcObj4=coda::mcmc(data=mu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu1)
# MU2
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu2,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[2]))
plot(mu2,ty="l",col="cornflowerblue",xlab=expression(mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=mu2)
mcmcObj2=coda::mcmc(data=mu2SecC)
mcmcObj3=coda::mcmc(data=mu2ThdC)
mcmcObj4=coda::mcmc(data=mu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj,auto.layout = FALSE)
stats::acf(mu2)
# MU3
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu3,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[3]))
plot(mu3,ty="l",col="cornflowerblue",xlab=expression(mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=mu3)
mcmcObj2=coda::mcmc(data=mu3SecC)
mcmcObj3=coda::mcmc(data=mu3ThdC)
mcmcObj4=coda::mcmc(data=mu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj,auto.layout = FALSE)
stats::acf(mu3)
# MU4
if(!is.null(dataframe$mu4)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu4,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[4]))
plot(mu4,ty="l",col="cornflowerblue",xlab=expression(mu[4]),ylab="")
mcmcObj1=coda::mcmc(data=mu4)
mcmcObj2=coda::mcmc(data=mu4SecC)
mcmcObj3=coda::mcmc(data=mu4ThdC)
mcmcObj4=coda::mcmc(data=mu4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu4)
}
# MU5
if(!is.null(dataframe$mu5)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu5,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[5]))
plot(mu5,ty="l",col="cornflowerblue",xlab=expression(mu[5]),ylab="")
mcmcObj1=coda::mcmc(data=mu5)
mcmcObj2=coda::mcmc(data=mu5SecC)
mcmcObj3=coda::mcmc(data=mu5ThdC)
mcmcObj4=coda::mcmc(data=mu5FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu5)
}
# MU6
if(!is.null(dataframe$mu6)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(mu6,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(mu[6]))
plot(mu6,ty="l",col="cornflowerblue",xlab=expression(mu[6]),ylab="")
mcmcObj1=coda::mcmc(data=mu6)
mcmcObj2=coda::mcmc(data=mu6SecC)
mcmcObj3=coda::mcmc(data=mu6ThdC)
mcmcObj4=coda::mcmc(data=mu6FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(mu6)
}
# SD1
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma1Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[1]^2))
plot(sigma1Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(sigma1Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[1]))
plot(sigma1Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma1Sq)
mcmcObj2=coda::mcmc(data=sigma1SqSecC)
mcmcObj3=coda::mcmc(data=sigma1SqThdC)
mcmcObj4=coda::mcmc(data=sigma1SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma1Sq)
# SD2
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma2Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[2]^2))
plot(sigma2Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(sigma2Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[2]))
plot(sigma2Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma2Sq)
mcmcObj2=coda::mcmc(data=sigma2SqSecC)
mcmcObj3=coda::mcmc(data=sigma2SqThdC)
mcmcObj4=coda::mcmc(data=sigma2SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma2Sq)
# SD3
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma3Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[3]^2))
plot(sigma3Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(sigma3Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[3]))
plot(sigma3Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma3Sq)
mcmcObj2=coda::mcmc(data=sigma3SqSecC)
mcmcObj3=coda::mcmc(data=sigma3SqThdC)
mcmcObj4=coda::mcmc(data=sigma3SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma3Sq)
# SD4
if(!is.null(dataframe$mu4)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma4Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[4]^2))
plot(sigma4Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2),ylab="")
}
if(sd=="sd"){
hist(sigma4Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[4]))
plot(sigma4Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[4]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma4Sq)
mcmcObj2=coda::mcmc(data=sigma4SqSecC)
mcmcObj3=coda::mcmc(data=sigma4SqThdC)
mcmcObj4=coda::mcmc(data=sigma4SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma4Sq)
}
# SD5
if(!is.null(dataframe$mu5)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma5Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[5]^2))
plot(sigma5Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2),ylab="")
}
if(sd=="sd"){
hist(sigma5Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[5]))
plot(sigma5Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[5]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma5Sq)
mcmcObj2=coda::mcmc(data=sigma5SqSecC)
mcmcObj3=coda::mcmc(data=sigma5SqThdC)
mcmcObj4=coda::mcmc(data=sigma5SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma5Sq)
}
# SD6
if(!is.null(dataframe$mu6)){
dev.new();
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(sigma6Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[6]^2))
plot(sigma6Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2),ylab="")
}
if(sd=="sd"){
hist(sigma6Sq,freq=FALSE,main="",col="cornflowerblue",border="white",xlab=expression(sigma[6]))
plot(sigma6Sq,ty="l",col="cornflowerblue",xlab=expression(sigma[6]),ylab="")
}
mcmcObj1=coda::mcmc(data=sigma6Sq)
mcmcObj2=coda::mcmc(data=sigma6SqSecC)
mcmcObj3=coda::mcmc(data=sigma6SqThdC)
mcmcObj4=coda::mcmc(data=sigma6SqFrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(sigma6Sq)
}
}
# posterior plot of difference of means & variances
if(type=="diff"){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu2MinusMu1,freq=FALSE,main="Difference of means Group2-Group1",col="cornflowerblue",border="white",xlab=expression(mu[2]-mu[1]))
plot(diffOfMeansMu2MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[2]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu2MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu2MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu2MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu2MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu2MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu3MinusMu2,freq=FALSE,main="Difference of means Group3-Group2",col="cornflowerblue",border="white",xlab=expression(mu[3]-mu[2]))
plot(diffOfMeansMu3MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[3]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu3MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu3MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu3MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu3MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu3MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu3MinusMu1,freq=FALSE,main="Difference of means Group3-Group1",col="cornflowerblue",border="white",xlab=expression(mu[3]-mu[1]))
plot(diffOfMeansMu3MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[3]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu3MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu3MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu3MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu3MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu3MinusMu1)
if(!is.null(dataframe$mu4)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu4MinusMu1,freq=FALSE,main="Difference of means Group4-Group1",col="cornflowerblue",border="white",xlab=expression(mu[4]-mu[1]))
plot(diffOfMeansMu4MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[4]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu4MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu4MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu4MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu4MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu4MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu4MinusMu2,freq=FALSE,main="Difference of means Group2-Group2",col="cornflowerblue",border="white",xlab=expression(mu[4]-mu[2]))
plot(diffOfMeansMu4MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[4]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu4MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu4MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu4MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu4MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu4MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu4MinusMu3,freq=FALSE,main="Difference of means Group4-Group3",col="cornflowerblue",border="white",xlab=expression(mu[4]-mu[3]))
plot(diffOfMeansMu4MinusMu3,ty="l",col="cornflowerblue",xlab=expression(mu[4]-mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu4MinusMu3)
mcmcObj2=coda::mcmc(data=diffOfMeansMu4MinusMu3SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu4MinusMu3ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu4MinusMu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu4MinusMu3)
}
if(!is.null(dataframe$mu5)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu1,freq=FALSE,main="Difference of means Group5-Group1",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[1]))
plot(diffOfMeansMu5MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu2,freq=FALSE,main="Difference of means Group5-Group2",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[2]))
plot(diffOfMeansMu5MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu3,freq=FALSE,main="Difference of means Group5-Group3",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[3]))
plot(diffOfMeansMu5MinusMu3,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu3)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu3SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu3ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu5MinusMu4,freq=FALSE,main="Difference of means Group5-Group4",col="cornflowerblue",border="white",xlab=expression(mu[5]-mu[4]))
plot(diffOfMeansMu5MinusMu4,ty="l",col="cornflowerblue",xlab=expression(mu[5]-mu[4]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu5MinusMu4)
mcmcObj2=coda::mcmc(data=diffOfMeansMu5MinusMu4SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu5MinusMu4ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu5MinusMu4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu5MinusMu4)
}
if(!is.null(dataframe$mu6)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu1,freq=FALSE,main="Difference of means Group6-Group1",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[1]))
plot(diffOfMeansMu6MinusMu1,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[1]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu1)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu1SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu1ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu2,freq=FALSE,main="Difference of means Group6-Group2",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[2]))
plot(diffOfMeansMu6MinusMu2,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[2]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu2)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu2SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu2ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu3,freq=FALSE,main="Difference of means Group6-Group3",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[3]))
plot(diffOfMeansMu6MinusMu3,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[3]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu3)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu3SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu3ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu4,freq=FALSE,main="Difference of means Group6-Group4",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[4]))
plot(diffOfMeansMu6MinusMu4,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[4]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu4)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu4SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu4ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu4)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(diffOfMeansMu6MinusMu5,freq=FALSE,main="Difference of means Group6-Group5",col="cornflowerblue",border="white",xlab=expression(mu[6]-mu[5]))
plot(diffOfMeansMu6MinusMu5,ty="l",col="cornflowerblue",xlab=expression(mu[6]-mu[5]),ylab="")
mcmcObj1=coda::mcmc(data=diffOfMeansMu6MinusMu5)
mcmcObj2=coda::mcmc(data=diffOfMeansMu6MinusMu5SecC)
mcmcObj3=coda::mcmc(data=diffOfMeansMu6MinusMu5ThdC)
mcmcObj4=coda::mcmc(data=diffOfMeansMu6MinusMu5FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfMeansMu6MinusMu5)
}
# DIFFERENCES OF VARIANCES / STANDARD DEVIATIONS
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS2MinusS1 ,freq=FALSE,main="Difference of variances Group2-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[2]^2-sigma[1]^2))
plot(diffOfVariancesS2MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[2]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS2MinusS1,freq=FALSE,main="Difference of standard deviations Group2-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[2]-sigma[1]))
plot(diffOfVariancesS2MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[2]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS2MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS2MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS2MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS2MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS2MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS3MinusS2,freq=FALSE,main="Difference of variances Group3-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[3]^2-sigma[2]^2))
plot(diffOfVariancesS3MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[3]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS3MinusS2,freq=FALSE,main="Difference of standard deviations Group3-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[3]-sigma[2]))
plot(diffOfVariancesS3MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[3]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS3MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS3MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS3MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS3MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS3MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS3MinusS1,freq=FALSE,main="Difference of variances Group3-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[3]^2-sigma[1]^2))
plot(diffOfVariancesS3MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[3]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS3MinusS1,freq=FALSE,main="Difference of standard deviations Group3-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[3]-sigma[1]))
plot(diffOfVariancesS3MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[3]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS3MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS3MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS3MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS3MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS3MinusS1)
if(!is.null(dataframe$mu4)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS4MinusS1,freq=FALSE,main="Difference of variances Group4-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[4]^2-sigma[1]^2))
plot(diffOfVariancesS4MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS4MinusS1,freq=FALSE,main="Difference of standard deviations Group4-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[4]-sigma[1]))
plot(diffOfVariancesS4MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[4]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS4MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS4MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS4MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS4MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS4MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS4MinusS2,freq=FALSE,main="Difference of variances Group4-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[4]^2-sigma[2]^2))
plot(diffOfVariancesS4MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS4MinusS2,freq=FALSE,main="Difference of standard deviations Group4-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[4]-sigma[2]))
plot(diffOfVariancesS4MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[4]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS4MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS4MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS4MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS4MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS4MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS4MinusS3,freq=FALSE,main="Difference of variances Group4-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[4]^2-sigma[3]^2))
plot(diffOfVariancesS4MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[4]^2-sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS4MinusS3,freq=FALSE,main="Difference of standard deviations Group4-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[4]-sigma[3]))
plot(diffOfVariancesS4MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[4]-sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS4MinusS3)
mcmcObj2=coda::mcmc(data=diffOfVariancesS4MinusS3SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS4MinusS3ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS4MinusS3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS4MinusS3)
}
if(!is.null(dataframe$mu5)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS1,freq=FALSE,main="Difference of variances Group5-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[1]^2))
plot(diffOfVariancesS5MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS1,freq=FALSE,main="Difference of standard deviations Group5-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[1]))
plot(diffOfVariancesS5MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS2,freq=FALSE,main="Difference of variances Group5-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[2]^2))
plot(diffOfVariancesS5MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS2,freq=FALSE,main="Difference of standard deviations Group5-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[2]))
plot(diffOfVariancesS5MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS3,freq=FALSE,main="Difference of variances Group5-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[3]^2))
plot(diffOfVariancesS5MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS3,freq=FALSE,main="Difference of standard deviations Group5-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[3]))
plot(diffOfVariancesS5MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS3)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS3SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS3ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS5MinusS4,freq=FALSE,main="Difference of variances Group5-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[4]^2))
plot(diffOfVariancesS5MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[5]^2-sigma[4]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS5MinusS4,freq=FALSE,main="Difference of standard deviations Group5-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[5]-sigma[4]))
plot(diffOfVariancesS5MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[5]-sigma[4]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS5MinusS4)
mcmcObj2=coda::mcmc(data=diffOfVariancesS5MinusS4SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS5MinusS4ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS5MinusS4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS5MinusS4)
}
if(!is.null(dataframe$mu6)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS1,freq=FALSE,main="Difference of variances Group6-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[1]^2))
plot(diffOfVariancesS6MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[1]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS1,freq=FALSE,main="Difference of standard deviations Group6-Group1",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[1]))
plot(diffOfVariancesS6MinusS1,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[1]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS1)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS1SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS1ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS1FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS1)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS2,freq=FALSE,main="Difference of variances Group6-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[2]^2))
plot(diffOfVariancesS6MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[2]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS2,freq=FALSE,main="Difference of standard deviations Group6-Group2",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[2]))
plot(diffOfVariancesS6MinusS2,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[2]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS2)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS2SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS2ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS2FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS2)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS3,freq=FALSE,main="Difference of variances Group6-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[3]^2))
plot(diffOfVariancesS6MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[3]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS3,freq=FALSE,main="Difference of standard deviations Group6-Group3",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[3]))
plot(diffOfVariancesS6MinusS3,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[3]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS3)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS3SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS3ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS3FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS3)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS4,freq=FALSE,main="Difference of variances Group6-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[4]^2))
plot(diffOfVariancesS6MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[4]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS4,freq=FALSE,main="Difference of standard deviations Group6-Group4",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[4]))
plot(diffOfVariancesS6MinusS4,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[4]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS4)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS4SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS4ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS4FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS4)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
if(sd=="var"){
hist(diffOfVariancesS6MinusS5,freq=FALSE,main="Difference of variances Group6-Group5",col="cornflowerblue",border="white",
xlab=expression(sigma[5]^2-sigma[5]^2))
plot(diffOfVariancesS6MinusS5,ty="l",col="cornflowerblue",xlab=expression(sigma[6]^2-sigma[5]^2),ylab="")
}
if(sd=="sd"){
hist(diffOfVariancesS6MinusS5,freq=FALSE,main="Difference of standard deviations Group6-Group5",col="cornflowerblue",border="white",
xlab=expression(sigma[6]-sigma[5]))
plot(diffOfVariancesS6MinusS5,ty="l",col="cornflowerblue",xlab=expression(sigma[6]-sigma[5]),ylab="")
}
mcmcObj1=coda::mcmc(data=diffOfVariancesS6MinusS5)
mcmcObj2=coda::mcmc(data=diffOfVariancesS6MinusS5SecC)
mcmcObj3=coda::mcmc(data=diffOfVariancesS6MinusS5ThdC)
mcmcObj4=coda::mcmc(data=diffOfVariancesS6MinusS5FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(diffOfVariancesS6MinusS5)
}
}
# posterior plot of effect size
if(type=="effect"){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize12,freq=FALSE,main="Effect size Group1-Group2",col="cornflowerblue",border="white",xlab=expression(delta[12]))
plot(effectSize12,ty="l",col="cornflowerblue",xlab=expression(delta[12]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize12)
mcmcObj2=coda::mcmc(data=effectSize12SecC)
mcmcObj3=coda::mcmc(data=effectSize12ThdC)
mcmcObj4=coda::mcmc(data=effectSize12FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize12)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize23,freq=FALSE,main="Effect size Group2-Group3",col="cornflowerblue",border="white",xlab=expression(delta[23]))
plot(effectSize23,ty="l",col="cornflowerblue",xlab=expression(delta[23]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize23)
mcmcObj2=coda::mcmc(data=effectSize23SecC)
mcmcObj3=coda::mcmc(data=effectSize23ThdC)
mcmcObj4=coda::mcmc(data=effectSize23FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize23)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize13,freq=FALSE,main="Effect size Group1-Group3",col="cornflowerblue",border="white",xlab=expression(delta[13]))
plot(effectSize13,ty="l",col="cornflowerblue",xlab=expression(delta[13]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize13)
mcmcObj2=coda::mcmc(data=effectSize13SecC)
mcmcObj3=coda::mcmc(data=effectSize13ThdC)
mcmcObj4=coda::mcmc(data=effectSize13FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize13)
if(!is.null(dataframe$mu4)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize14,freq=FALSE,main="Effect size Group1-Group4",col="cornflowerblue",border="white",xlab=expression(delta[14]))
plot(effectSize14,ty="l",col="cornflowerblue",xlab=expression(delta[14]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize14)
mcmcObj2=coda::mcmc(data=effectSize14SecC)
mcmcObj3=coda::mcmc(data=effectSize14ThdC)
mcmcObj4=coda::mcmc(data=effectSize14FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize14)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize24,freq=FALSE,main="Effect size Group2-Group4",col="cornflowerblue",border="white",xlab=expression(delta[24]))
plot(effectSize24,ty="l",col="cornflowerblue",xlab=expression(delta[24]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize24)
mcmcObj2=coda::mcmc(data=effectSize24SecC)
mcmcObj3=coda::mcmc(data=effectSize24ThdC)
mcmcObj4=coda::mcmc(data=effectSize24FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize24)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize34,freq=FALSE,main="Effect size Group3-Group4",col="cornflowerblue",border="white",xlab=expression(delta[34]))
plot(effectSize34,ty="l",col="cornflowerblue",xlab=expression(delta[34]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize34)
mcmcObj2=coda::mcmc(data=effectSize34SecC)
mcmcObj3=coda::mcmc(data=effectSize34ThdC)
mcmcObj4=coda::mcmc(data=effectSize34FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize34)
}
if(!is.null(dataframe$mu5)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize15,freq=FALSE,main="Effect size Group1-Group5",col="cornflowerblue",border="white",xlab=expression(delta[15]))
plot(effectSize15,ty="l",col="cornflowerblue",xlab=expression(delta[15]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize15)
mcmcObj2=coda::mcmc(data=effectSize15SecC)
mcmcObj3=coda::mcmc(data=effectSize15ThdC)
mcmcObj4=coda::mcmc(data=effectSize15FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize15)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize25,freq=FALSE,main="Effect size Group2-Group5",col="cornflowerblue",border="white",xlab=expression(delta[25]))
plot(effectSize25,ty="l",col="cornflowerblue",xlab=expression(delta[25]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize25)
mcmcObj2=coda::mcmc(data=effectSize25SecC)
mcmcObj3=coda::mcmc(data=effectSize25ThdC)
mcmcObj4=coda::mcmc(data=effectSize25FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize25)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize35,freq=FALSE,main="Effect size Group3-Group5",col="cornflowerblue",border="white",xlab=expression(delta[35]))
plot(effectSize35,ty="l",col="cornflowerblue",xlab=expression(delta[35]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize35)
mcmcObj2=coda::mcmc(data=effectSize35SecC)
mcmcObj3=coda::mcmc(data=effectSize35ThdC)
mcmcObj4=coda::mcmc(data=effectSize35FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize35)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize45,freq=FALSE,main="Effect size Group4-Group5",col="cornflowerblue",border="white",xlab=expression(delta[45]))
plot(effectSize45,ty="l",col="cornflowerblue",xlab=expression(delta[45]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize45)
mcmcObj2=coda::mcmc(data=effectSize45SecC)
mcmcObj3=coda::mcmc(data=effectSize45ThdC)
mcmcObj4=coda::mcmc(data=effectSize45FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize45)
}
if(!is.null(dataframe$mu6)){
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize16,freq=FALSE,main="Effect size Group1-Group6",col="cornflowerblue",border="white",xlab=expression(delta[16]))
plot(effectSize16,ty="l",col="cornflowerblue",xlab=expression(delta[16]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize16)
mcmcObj2=coda::mcmc(data=effectSize16SecC)
mcmcObj3=coda::mcmc(data=effectSize16ThdC)
mcmcObj4=coda::mcmc(data=effectSize16FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize16)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize26,freq=FALSE,main="Effect size Group2-Group6",col="cornflowerblue",border="white",xlab=expression(delta[26]))
plot(effectSize26,ty="l",col="cornflowerblue",xlab=expression(delta[26]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize26)
mcmcObj2=coda::mcmc(data=effectSize26SecC)
mcmcObj3=coda::mcmc(data=effectSize26ThdC)
mcmcObj4=coda::mcmc(data=effectSize26FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize26)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize36,freq=FALSE,main="Effect size Group3-Group6",col="cornflowerblue",border="white",xlab=expression(delta[36]))
plot(effectSize36,ty="l",col="cornflowerblue",xlab=expression(delta[36]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize36)
mcmcObj2=coda::mcmc(data=effectSize36SecC)
mcmcObj3=coda::mcmc(data=effectSize36ThdC)
mcmcObj4=coda::mcmc(data=effectSize36FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize36)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize46,freq=FALSE,main="Effect size Group4-Group6",col="cornflowerblue",border="white",xlab=expression(delta[46]))
plot(effectSize46,ty="l",col="cornflowerblue",xlab=expression(delta[46]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize46)
mcmcObj2=coda::mcmc(data=effectSize46SecC)
mcmcObj3=coda::mcmc(data=effectSize46ThdC)
mcmcObj4=coda::mcmc(data=effectSize46FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize46)
dev.new()
plotpar<-par(mfrow=c(2,2),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(plotpar))
hist(effectSize56,freq=FALSE,main="Effect size Group5-Group6",col="cornflowerblue",border="white",xlab=expression(delta[56]))
plot(effectSize56,ty="l",col="cornflowerblue",xlab=expression(delta[56]),ylab="")
mcmcObj1=coda::mcmc(data=effectSize56)
mcmcObj2=coda::mcmc(data=effectSize56SecC)
mcmcObj3=coda::mcmc(data=effectSize56ThdC)
mcmcObj4=coda::mcmc(data=effectSize56FrtC)
mcmcListObj=coda::mcmc.list(mcmcObj1,mcmcObj2,mcmcObj3,mcmcObj4)
coda::gelman.plot(mcmcListObj, auto.layout = FALSE)
stats::acf(effectSize56)
}
}
# ROPE posterior analysis plot
if(type=="rope"){
# utility function for computing the mode
getmode <- function(v) {
uniqv <- unique(v)
uniqv[which.max(tabulate(match(v, uniqv)))]
}
################# POSTERIOR ROPE ANALYSIS FOR DELTA_12 ######################
# Set 2x3 par
dev.new()
newpar <- par(mfrow=c(2,3),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize12,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[12]),main=mtext("Posterior distribution of Effect size "~delta[12], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize12),getmode(effectSize12)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize12,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize12),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize12),3)),side=3,line=1.15,at=min(effectSize12),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize12),3)),side=3,line=1.15,at=max(effectSize12),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_13 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize13,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[13]),main=mtext("Posterior distribution of Effect size "~delta[13], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize13),getmode(effectSize13)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize13,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize13),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize13),3)),side=3,line=1.15,at=min(effectSize13),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize13),3)),side=3,line=1.15,at=max(effectSize13),col="black",cex=0.8,font=2)
####################################### HISTOGRAM DELTA_23 ###############################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize23,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[23]),main=mtext("Posterior distribution of Effect size "~delta[23], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize23),getmode(effectSize23)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize23,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize23),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize23),3)),side=3,line=1.15,at=min(effectSize23),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize23),3)),side=3,line=1.15,at=max(effectSize23),col="black",cex=0.8,font=2)
######################## POSTERIOR ROPE ANALYSIS DELTA_12 #################################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize12,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize12[effectSize12 >effSizeCI[1] & effectSize12 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[12]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
################################## POSTERIOR ROPE ANALYSIS FOR DELTA_[13] ############################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize13,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize13[effectSize13 >effSizeCI[1] & effectSize13 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[13]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_23 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize23,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize23[effectSize23 >effSizeCI[1] & effectSize23 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[23]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
###################################################### ADDITIONAL EFFECT SIZES FOR FOURTH GROUP #########################################################
if(!is.null(dataframe$mu4)){
# Set 2x3 par
dev.new()
newpar <- par(mfrow=c(2,3),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
########################################### HISTOGRAM DELTA_14 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize14,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[14]),main=mtext("Posterior distribution of Effect size "~delta[14], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize14),getmode(effectSize14)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize14,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize14),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize14),3)),side=3,line=1.15,at=min(effectSize14),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize14),3)),side=3,line=1.15,at=max(effectSize14),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_24 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize24,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[24]),main=mtext("Posterior distribution of Effect size "~delta[24], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize24),getmode(effectSize24)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize24,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-2,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,2),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize24),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize24),3)),side=3,line=1.15,at=min(effectSize24),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize24),3)),side=3,line=1.15,at=max(effectSize24),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_34 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize34,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[34]),main=mtext("Posterior distribution of Effect size "~delta[34], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize34),getmode(effectSize34)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize34,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-2,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,2),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize34),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize34),3)),side=3,line=1.15,at=min(effectSize34),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize34),3)),side=3,line=1.15,at=max(effectSize34),col="black",cex=0.8,font=2)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_14 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize14,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize14[effectSize14 >effSizeCI[1] & effectSize14 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[14]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_24 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize24,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize24[effectSize24 >effSizeCI[1] & effectSize24 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[24]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_34 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize34,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize34[effectSize34 >effSizeCI[1] & effectSize34 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[34]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
}
###################################################### ADDITIONAL EFFECT SIZES FOR FIFTH GROUP #########################################################
if(!is.null(dataframe$mu5)){
# Set 2x4 par
dev.new()
newpar <- par(mfrow=c(2,4),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
########################################### HISTOGRAM DELTA_15 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize15,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[15]),main=mtext("Posterior distribution of Effect size "~delta[15], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize15),getmode(effectSize15)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize15,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize15),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize15),3)),side=3,line=1.15,at=min(effectSize15),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize15),3)),side=3,line=1.15,at=max(effectSize15),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_25 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize25,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[25]),main=mtext("Posterior distribution of Effect size "~delta[25], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize25),getmode(effectSize25)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize25,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize25),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize25),3)),side=3,line=1.15,at=min(effectSize25),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize25),3)),side=3,line=1.15,at=max(effectSize25),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_35 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize35,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[35]),main=mtext("Posterior distribution of Effect size "~delta[35], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize35),getmode(effectSize35)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize35,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize35),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize35),3)),side=3,line=1.15,at=min(effectSize35),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize35),3)),side=3,line=1.15,at=max(effectSize35),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_45 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize45,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[45]),main=mtext("Posterior distribution of Effect size "~delta[45], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize45),getmode(effectSize45)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize45,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize45),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize45),3)),side=3,line=1.15,at=min(effectSize45),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize45),3)),side=3,line=1.15,at=max(effectSize45),col="black",cex=0.8,font=2)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_15 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize15,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize15[effectSize15 >effSizeCI[1] & effectSize15 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[15]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_25 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize25,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize25[effectSize25 >effSizeCI[1] & effectSize25 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[25]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_35 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize35,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize35[effectSize35 >effSizeCI[1] & effectSize35 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[35]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_45 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize45,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize45[effectSize45 >effSizeCI[1] & effectSize45 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[45]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
}
###################################################### ADDITIONAL EFFECT SIZES FOR SIXTH GROUP #########################################################
if(!is.null(dataframe$mu6)){
# Set 2x4 par
dev.new()
newpar <- par(mfrow=c(2,5),mai = c(1,0.5,0.5,0.5), oma=c(1,1,1,1))
# Reset par only after first change of par
on.exit(par(newpar))
########################################### HISTOGRAM DELTA_16 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize16,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[16]),main=mtext("Posterior distribution of Effect size "~delta[16], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize16),getmode(effectSize16)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize16,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize16),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize16),3)),side=3,line=1.15,at=min(effectSize16),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize16),3)),side=3,line=1.15,at=max(effectSize16),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_26 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize26,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[26]),main=mtext("Posterior distribution of Effect size "~delta[26], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize26),getmode(effectSize26)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize26,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize26),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize26),3)),side=3,line=1.15,at=min(effectSize26),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize26),3)),side=3,line=1.15,at=max(effectSize26),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_36 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize36,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[36]),main=mtext("Posterior distribution of Effect size "~delta[36], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize36),getmode(effectSize36)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize36,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize36),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize36),3)),side=3,line=1.15,at=min(effectSize36),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize36),3)),side=3,line=1.15,at=max(effectSize36),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_46 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize46,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[46]),main=mtext("Posterior distribution of Effect size "~delta[46], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize46),getmode(effectSize46)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize46,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize46),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize46),3)),side=3,line=1.15,at=min(effectSize46),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize46),3)),side=3,line=1.15,at=max(effectSize46),col="black",cex=0.8,font=2)
########################################### HISTOGRAM DELTA_56 ###########################################
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(effectSize56,freq=FALSE,col="cornflowerblue",border="white",xlab=expression(delta[56]),main=mtext("Posterior distribution of Effect size "~delta[56], side=3,line=2.2,col="black",cex=0.8,font=2),xaxt="n")
ticks <- seq(from=-10, to=10, by=0.05)
# And draw the axis:
axis(1, at=ticks)
# Posterior mode
#abline(v=getmode(effectSize),col="blue",lwd=2,lty="solid")
lines(x=c(getmode(effectSize56),getmode(effectSize56)),y=c(-0.025,0.025),type="l",lwd=2.5,col="blue")
# Posterior CI
postCI=quantile(effectSize56,probs=c((1-ci)/2,ci+(1-ci)/2))
lines(x=c(postCI[1],postCI[2]),y=c(0,0),type="l",lwd=2,col="black")
lines(x=c(postCI[1],postCI[1]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
lines(x=c(postCI[2],postCI[2]),y=c(-0.02,0.02),type="l",lwd=2,col="black")
# Visualization of typical ROPEs
# ROPE for no effect
lines(x=c(-0.2,0.2),y=c(0.04,0.04),type="l",lwd=2,col="red")
# ROPE for small effect
lines(x=c(-0.5,-0.2),y=c(0.04,0.04),type="l",lwd=2,col="orange")
lines(x=c(0.2,0.5),y=c(0.04,0.04),type="l",lwd=2,col="orange")
# ROPE for medium effect
lines(x=c(-0.8,-0.5),y=c(0.04,0.04),type="l",lwd=2,col="green")
lines(x=c(0.5,0.8),y=c(0.04,0.04),type="l",lwd=2,col="green")
# ROPE for large effect
lines(x=c(-10,-0.8),y=c(0.04,0.04),type="l",lwd=2,col="purple")
lines(x=c(0.8,10),y=c(0.04,0.04),type="l",lwd=2,col="purple")
abline(v=0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=-0.8,col="purple",lwd=0.5,lty="dashed")
abline(v=0.5,col="green",lwd=0.5,lty="dashed")
abline(v=-0.5,col="green",lwd=0.5,lty="dashed")
abline(v=0.2,col="orange",lwd=0.5,lty="dashed")
abline(v=-0.2,col="orange",lwd=0.5,lty="dashed")
# Positions of text above the plot
#posLeft=quantile(effectSize,probs=c(0.1,0.5,0.95))[1]
#posMid=quantile(effectSize,probs=c(0.1,0.5,0.95))[2]
#posRight=quantile(effectSize,probs=c(0.1,0.5,0.95))[3]
# Text above the plot
mtext(paste0(ci*100,"% ","CI: [", round(postCI[1], 3),",",round(postCI[2], 3),"]"), side=3,line=0.15,at=min(effectSize56),col="black",cex=0.8,font=2)
mtext(paste0("Mean: ", round(mean(effectSize56),3)),side=3,line=1.15,at=min(effectSize56),col="black",cex=0.8,font=2)
mtext(paste0("Mode: ", round(getmode(effectSize56),3)),side=3,line=1.15,at=max(effectSize56),col="black",cex=0.8,font=2)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_15 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize16,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize16[effectSize16 >effSizeCI[1] & effectSize16 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[16]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_26 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize26,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize26[effectSize26 >effSizeCI[1] & effectSize26 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
# Barplot
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[26]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_36 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize36,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize36[effectSize36 >effSizeCI[1] & effectSize36 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[36]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_46 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize46,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize46[effectSize46 >effSizeCI[1] & effectSize46 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[46]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
####################### POSTERIOR ROPE ANALYSIS FOR DELTA_56 #######################################
# Bar plot
#effSizeCIValues = effectSize[{q<-rank(effectSize)/length(effectSize);q>=((1-ci)/2) & q <= (ci+(1-ci)/2)}]
effSizeCI=quantile(effectSize56,probs=c(((1-ci)/2),(ci+(1-ci)/2)))
effectSizeCIValues = effectSize56[effectSize56 >effSizeCI[1] & effectSize56 < effSizeCI[2]]
largePosEffectIterations = length(which(effectSizeCIValues >= 0.8))/length(effectSizeCIValues)
largeNegEffectIterations = length(which(effectSizeCIValues <= -0.8))/length(effectSizeCIValues)
mediumPosEffectIterations = length(which(effectSizeCIValues >= 0.5 & effectSizeCIValues < 0.8))/length(effectSizeCIValues)
mediumNegEffectIterations = length(which(effectSizeCIValues <= -0.5 & effectSizeCIValues > -0.8))/length(effectSizeCIValues)
smallPosEffectIterations = length(which(effectSizeCIValues >= 0.2 & effectSizeCIValues < 0.5))/length(effectSizeCIValues)
smallNegEffectIterations = length(which(effectSizeCIValues <= -0.2 & effectSizeCIValues > -0.5))/length(effectSizeCIValues)
noEffectIterations = length(which(effectSizeCIValues < 0.2 & effectSizeCIValues > -0.2))/length(effectSizeCIValues)
lbls=c("large positive:", "medium positive:", "small positive:", "no effect:", "small negative:", "medium negative:","large negative:")
pct=c(largeNegEffectIterations,mediumNegEffectIterations,smallNegEffectIterations,noEffectIterations,smallPosEffectIterations,mediumPosEffectIterations,largePosEffectIterations)
pct=round(pct,digits=4)
pct=pct*100
lbls=paste(lbls,pct)
lbls <- paste(pct,"%",sep="") # ad % to labels
#pie(pct,labels = lbls, col=rainbow(length(lbls)),
#main="Posterior probabilities of varying effect sizes")
b<-barplot(pct,
main = mtext("Posterior ROPE-analysis for"~delta[56]),
xlab = "ROPEs of standard effect sizes",
ylab = "Posterior-Percentage included inside ROPE",
names.arg = c("L-", "M-", "S-", "No Eff.", "S+", "M+", "L+"),
col = "cornflowerblue",
border="white",
horiz = FALSE,
ylim=c(0,107.5))
b
text(x=b,y=pct+4,labels=lbls)
}
}
}
assumption.check = function(x1,x2,x3,x4=NULL,x5=NULL,x6=NULL,conf.level=0.95){
check=TRUE;
if(shapiro.test(x1)$p.value < 1-conf.level | shapiro.test(x2)$p.value < 1-conf.level | shapiro.test(x3)$p.value < 1-conf.level){
check=FALSE;
}
if(!is.null(x4)){
if(shapiro.test(x4)$p.value < 1-conf.level){
check=FALSE;
}
}
if(!is.null(x5)){
if(shapiro.test(x5)$p.value < 1-conf.level){
check=FALSE;
}
}
if(!is.null(x6)){
if(shapiro.test(x6)$p.value < 1-conf.level){
check=FALSE;
}
}
if(check==TRUE){
message("Model assumptions checked. No significant deviations from normality detected. Bayesian ANOVA can be run safely.")
} else {
warning("Model assumption of normally distributed data in each group is violated.\n All results of the Bayesian ANOVA based on a three-component Gaussian mixture could therefore be unreliable and not trustworthy.")
warning("Run further diagnostics (like Quantile-Quantile-plots) to check if the Bayesian ANOVA can be expected to be robust to the violations of normality\n")
}
# Set 2x3 par
dev.new()
newpar <- par(mfrow=c(2,3))
# Effect size posterior histogram
# Make a vector of values to draw ticks at:
hist(x1,freq=FALSE,col="cornflowerblue",border="white",xlab="First group",main="Histogram of first group")
lines(density(x1))
hist(x2,freq=FALSE,col="cornflowerblue",border="white",xlab="Second group",main="Histogram of second group")
lines(density(x2))
hist(x3,freq=FALSE,col="cornflowerblue",border="white",xlab="Third group",main="Histogram of third group")
lines(density(x3))
qqnorm(x1, col="cornflowerblue", main="Q-Q Plot for first group"); qqline(x1, col = "black")
qqnorm(x2, col="cornflowerblue", main="Q-Q Plot for second group"); qqline(x2, col = "black")
qqnorm(x3, col="cornflowerblue", main="Q-Q Plot for third group"); qqline(x3, col = "black")
if(length(x4)>1 && length(x5)>1 && length(x6)>1){
dev.new()
newpar <- par(mfrow=c(2,3))
hist(x4,freq=FALSE,col="cornflowerblue",border="white",xlab="Fourth group",main="Histogram of fourth group")
lines(density(x4))
hist(x5,freq=FALSE,col="cornflowerblue",border="white",xlab="Fifth group",main="Histogram of fifth group")
lines(density(x5))
hist(x6,freq=FALSE,col="cornflowerblue",border="white",xlab="Sixth group",main="Histogram of sixth group")
lines(density(x6))
qqnorm(x4, col="cornflowerblue", main="Q-Q Plot for fourth group"); qqline(x4, col = "black")
qqnorm(x5, col="cornflowerblue", main="Q-Q Plot for fifth group"); qqline(x5, col = "black")
qqnorm(x6, col="cornflowerblue", main="Q-Q Plot for sixth group"); qqline(x6, col = "black")
} else if(length(x4)>1 && length(x5)>1 && is.null(x6)){
dev.new()
newpar <- par(mfrow=c(2,2))
hist(x4,freq=FALSE,col="cornflowerblue",border="white",xlab="Fourth group",main="Histogram of fourth group")
lines(density(x4))
hist(x5,freq=FALSE,col="cornflowerblue",border="white",xlab="Fifth group",main="Histogram of fifth group")
lines(density(x5))
qqnorm(x4, col="cornflowerblue", main="Q-Q Plot for fourth group"); qqline(x4, col = "black")
qqnorm(x5, col="cornflowerblue", main="Q-Q Plot for fifth group"); qqline(x5, col = "black")
} else if(length(x4)>1 && is.null(x5) && is.null(x6)){
dev.new()
newpar <- par(mfrow=c(2,1))
hist(x4,freq=FALSE,col="cornflowerblue",border="white",xlab="Fourth group",main="Histogram of fourth group")
lines(density(x4))
qqnorm(x4, col="cornflowerblue", main="Q-Q Plot for fourth group"); qqline(x4, col = "black")
}
}
post.pred.check <- function(anovafit, ngroups, out, reps = 50, eta){
plot(density(out),main="Posterior predictive check",ylim=c(0,max(density(out)$y)*1.5))
for(i in 1:reps){
mu1 = sample(anovafit$mu1,size = 1)
mu2 = sample(anovafit$mu2,size = 1)
mu3 = sample(anovafit$mu3,size = 1)
sigma1Sq = sample(anovafit$sigma1Sq,size = 1)
sigma2Sq = sample(anovafit$sigma2Sq,size = 1)
sigma3Sq = sample(anovafit$sigma3Sq,size = 1)
if(ngroups > 3){
mu4 = sample(anovafit$mu4,size = 1)
sigma4Sq = sample(anovafit$sigma4Sq,size = 1)
}
if(ngroups > 4){
mu5 = sample(anovafit$mu5,size = 1)
sigma5Sq = sample(anovafit$sigma5Sq,size = 1)
}
if(ngroups > 5){
mu6 = sample(anovafit$mu6,size = 1)
sigma6Sq = sample(anovafit$sigma6Sq,size = 1)
}
x = seq(min(out),max(out),length.out = 50)
if(ngroups == 3){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
if(ngroups == 4){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq) +
eta[4]*dnorm(x,mu4,sigma4Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
if(ngroups == 5){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq) +
eta[4]*dnorm(x,mu4,sigma4Sq) +
eta[5]*dnorm(x,mu5,sigma5Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
if(ngroups == 6){
lines(x, eta[1]*dnorm(x,mu1,sigma1Sq) +
eta[2]*dnorm(x,mu2,sigma2Sq) +
eta[3]*dnorm(x,mu3,sigma3Sq) +
eta[4]*dnorm(x,mu4,sigma4Sq) +
eta[5]*dnorm(x,mu5,sigma5Sq) +
eta[6]*dnorm(x,mu6,sigma6Sq),
col = rgb(red = 0, green = 0, blue = 1, alpha = 0.2))
}
}
}
Try the bayesanova package in your browser
Any scripts or data that you put into this service are public.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.