ed.jags: Empirical determinacy for JAGS

In hunansona/ed4bhm: Empirical Determinacy for Bayesian Hierarchical Models

Empirical determinacy for JAGS

Description

Calculates the total empirical determinacy (TED) and proportional empirical determinacy for location (pEDL) and for spread (pEDS) for JAGS.

Usage

ed.jags(jags.object.base, distance = "H2ALL", delta = 0.01, prec.name)

Arguments

jags.object.base	character string, the name of the JAGS object fitted with rjags with the base model (non-weighted likelihood)
distance	character string, specifies the type of the empirical determinacy measure. It can have values "H2" for TED, "BCL" for EDL and "BCS" for EDS and "H2ALL" for all the five measures (TED, EDL, EDS, pEDL and pEDS). The default value is "H2ALL".
delta	numeric, numerical differentiation step, the weighting factor w = 1 \pm δ, the default value is 0.01.
prec.name	character string, the name of the precision parameter to be used for logarithmic transformation

Details

The first element in the final returned value does not depend on the value of distance. The argument distance = "H2ALL" provides a list, where the second element of the list is a matrix of five empirical determinacy measures.

Value

A list of length two. The first element of the list is a list containing the matrix of descriptive statistics (mean and standard deviation) for each parameter in the model and the value of δ. The second element of the list is a matrix of empirical determinacy measures for each parameter in the model. This matrix has columns c("TED", "EDL", "EDS", "pEDL", "pEDS") when distance = "H2ALL".

References

Hunanyan, S., Roos, M., Plummer, M., Rue, H. (2021). Quantification of empirical determinacy: the impact of likelihood weighting on posterior location and spread in Bayesian meta-analysis estimated with JAGS and INLA. Bayesian Analaysis (under review). https://arxiv.org/abs/2109.11870.

See Also

ed, package rjags

Examples

data(eight_schools)
#prior settings
mean_mu<-0
prec_mu<-1/(4^2) 
prec_tau<-1/(5^2)

library(rjags)
#sessionInfo()
list.modules()
# In order to sample from the likelihood in JAGS one has to load the "dic" module and 
#mention "deviance" in variable.names
load.module("dic")
list.modules()

set.seed(44566)

###############
# rjags interface with code in a model string
###############

# In JAGS the normal distribution is parametrized by its mean and precision
jags_data<-list(y=eight_schools$y, sigma_2=eight_schools$prec, 
                mean_mu=mean_mu, prec_mu=prec_mu, 
                prec_tau=prec_tau)

# define parameters
schools.params.jags<-c("mu", "tau_2", "deviance")

schools_centered_modelString <- "
# likelihood
model{
for(i in 1:length(y)){
y[i] ~ dnorm(theta[i], sigma_2[i])
theta[i] ~ dnorm(mu, tau_2);
}

# priors
mu ~ dnorm(mean_mu, prec_mu);
tau ~ dnorm(0,prec_tau)T(0,);
tau_2 <- 1/(tau^2);
}
"

writeLines(schools_centered_modelString, con="TempModel.txt") # write to a file

# # model initiation
rjags.8schools  <- jags.model(
  file = "TempModel.txt",
  data = jags_data,
  n.chains = 4,
  n.adapt = 4000,
  inits = list( list(mu = 3.1, tau = 3.117181, 
                .RNG.name = "base::Wichmann-Hill", .RNG.seed = 314159),
                list(mu = 3.3, tau = 2.517181, 
                .RNG.name = "base::Marsaglia-Multicarry", .RNG.seed = 159314),
                list(mu = 3.6, tau = 3.317181, 
                .RNG.name = "base::Super-Duper", .RNG.seed = 413159),
                list(mu = 3.8, tau = 2.317181,  
                .RNG.name = "base::Mersenne-Twister", .RNG.seed = 143915)))

# burn-in
update(rjags.8schools , n.iter = 20000)
N.iter <- 2*(10^5)
n.thin <- 100
# sampling/monitoring
fit.rjags.8schools  <- coda.samples(
  model = rjags.8schools ,
  variable.names = schools.params.jags,
  n.iter = N.iter,
  thin = n.thin)
  
ed_jags_8schools <- ed.jags(jags.object.base = fit.rjags.8schools, distance = "H2ALL",
                            delta = 0.01, prec.name="tau_2")

hunansona/ed4bhm documentation built on June 15, 2022, 6:42 p.m.