HSI
From: Bayesian Models for Astrophysical Data, Cambridge Univ. Press
(c) 2017, Joseph M. Hilbe, Rafael S. de Souza and Emille E. O. Ishida
you are kindly asked to include the complete citation if you used this material in a publication
Code 5.33 Explicitly given beta–binomial data in R
========================================================
y <- c(6,11,9,13,17,21,8,10,15,19,7,12)
m <- c(45,54,39,47,29,44,36,57,62,55,66,48)
x1 <- c(1,1,1,1,1,1,0,0,0,0,0,0)
x2 <- c(1,1,0,0,1,1,0,0,1,1,0,0)
x3 <- c(1,0,1,0,1,0,1,0,1,0,1,0)
bindata <-data.frame(y,m,x1,x2,x3)
========================================================
Code 5.34 Beta–binomial model in R using JAGS for explicitly given data and the zero trick
=======================================================================
library(R2jags)
X <- model.matrix(~ x1 + x2 + x3, data = bindata)
K <- ncol(X)
model.data <- list(Y = bindata$y,
N = nrow(bindata),
X =X,
K =K,
m = m,
Zeros = rep(0, nrow(bindata))
)
sink("BBL.txt")
cat("
model{
# Diffuse normal priors betas
for (i in 1:K) { beta[i] ~ dnorm(0, 0.0001)}
# Prior for sigma
sigma ~ dunif(0, 100)
C <- 10000
for (i in 1:N){
Zeros[i] ~ dpois(Zeros.mean[i])
Zeros.mean[i] <- -LL[i] + C
#mu[i] <- 1/(1+exp(-eta[i])) # can use for logit(mu[i]) below
logit(mu[i]) <- max(-20, min(20, eta[i]))
L1[i] <- loggam(m[i]+1) - loggam(Y[i]+1) - loggam(m[i]-Y[i]+1)
L2[i] <- loggam(1/sigma) + loggam(Y[i]+mu[i]/sigma)
L3[i] <- loggam(m[i] - Y[i]+(1-mu[i])/sigma) - loggam(m[i]+1/sigma)
L4[i] <- loggam(mu[i]/sigma) + loggam((1-mu[i])/sigma)
LL[i] <- L1[i] + L2[i] + L3[i] - L4[i]
eta[i] <- inprod(beta[], X[i,])
}
}
",fill = TRUE)
sink()
inits <- function () {list(beta = rnorm(K, 0, 0.1)) }
params <- c("beta", "sigma")
BBIN0 <- jags(data = model.data,
inits = inits,
parameters = params,
model.file = "BBL.txt",
n.thin = 3,
n.chains = 3,
n.burnin = 10000,
n.iter = 15000)
print(BBIN0, intervals=c(0.025, 0.975), digits=3)
=======================================================================
Output on screen:
Inference for Bugs model at "BBL.txt", fit using jags,
3 chains, each with 15000 iterations (first 10000 discarded),
n.thin = 3 n.sims = 5001 iterations saved
mu.vect sd.vect 2.5% 97.5% Rhat n.eff
beta[1] -1.222 0.416 -2.055 -0.375 1.001 4800
beta[2] 0.226 0.478 -0.725 1.161 1.001 5000
beta[3] 0.433 0.470 -0.506 1.349 1.001 5000
beta[4] -0.249 0.436 -1.143 0.622 1.001 2700
sigma 0.106 0.076 0.021 0.316 1.001 5000
deviance 240079.042 4.471 240073.054 240089.825 1.000 1
For each parameter, n.eff is a crude measure of effective sample size,
and Rhat is the potential scale reduction factor (at convergence, Rhat=1).
DIC info (using the rule, pD = var(deviance)/2)
pD = 10.0 and DIC = 240089.0
DIC is an estimate of expected predictive error (lower deviance is better).