criteria: criteria for comparison the bayesian beta regression

In Bayesianbetareg: Bayesian Beta regression: joint mean and precision modeling

Description

Performs the comparison criterias for the Bayesian Beta Regression

Usage

criteria(X,beta.residuals)

Arguments

X	object of class matrix, with the independent variable for the mean
beta.residuals	object of class bayesbetareg, with the residuals of the Bayesian Beta regression, that can be calculated by the function beta.residuals

Details

This function calculate the residuals of a bayesian beta regression.

Value

deviance	the deviance criteria
AIC	the AiC criteria
BIC	the BIC criteria

Author(s)

Daniel Jaimes [e-mail], Margarita Marin [e-mail], Javier Rojas [e-mail], Martha Corrales [e-mail] Maria Fernanda Zarate [e-mail] Ricardo Duplat [e-mail] Luis Villaraga [e-mail] Edilberto Cepeda-Cuervo [e-mail],

Examples

library(betareg)
data(ReadingSkills)

Y <- as.matrix(ReadingSkills[,1])
n <- length(Y)
X1 <- as.matrix(ReadingSkills[,2])
for(i in 1:length(X1)){
  X1 <- replace(X1,X1=="yes",1)
  X1 <- replace(X1,X1=="no",0)
}
X0 <- rep(1, times=n)
X1 <- as.numeric(X1)
X2 <- as.matrix(ReadingSkills[,3])
X3 <- X1*X2
X <- cbind(X0,X1,X2,X3)
Z0 <-  X0 
Z <- cbind(X0,X1)

burn <- 0.3
jump <- 3
nsim <- 400

bpri <- c(0,0,0,0)
Bpri <- diag(100,nrow=ncol(X),ncol=ncol(X))
gpri <- c(0,0)
Gpri <- diag(10,nrow=ncol(Z),ncol=ncol(Z))

re<-Bayesianbetareg(Y,X,Z,nsim,bpri,Bpri,gpri,Gpri,0.3,3,graph1=FALSE,graph2=FALSE)
summary(re)

readingskillsresiduals<- betaresiduals(Y,X,re)

readingskillscriterias <- criteria(X,readingskillsresiduals)

Bayesianbetareg documentation built on May 30, 2017, 2:35 a.m.