brand score estimation toy model with bayesian logistic regression

brand2.bugs

model {
  for (i in 1:N) {
    y[i] ~ dbern(p[i])
    for (j in 1:K) {
      q[i,j] <- b[j]*x[i,j]
    }
    logit(p[i]) <- a+sum(q[i,1:K])
  }
  a ~ dnorm(a_mu,a_tau)
  b[1] ~ dnorm(0,1.0)
  b[2] ~ dnorm(0,1.0)
}

brand2.R

library(R2WinBUGS)
library(coda)

setwd("~/R")

eps <- 1.0E-4
N <- 10
K <- 2

data1 <- list(
  x=cbind(
    c(100,120,90,100,80,70,200,100,150,110),
    c(0,1,0,1,0,0,1,0,0,0)
  ),
  y=c(0,1,0,1,0,1,1,1,0,0),
  a_mu=0.5,
  a_tau=1.0,
  N=N,K=K)

data2 <- list(
  x=cbind(
    c(100,120,90,100,80,70,200,100,150,110),
    c(0,1,0,1,0,0,1,0,0,0)
  ),
  y=c(1,0,1,0,1,1,0,0,1,1),
  a_mu=0.5,
  a_tau=1.0,
  N=N,K=K)

inits <- function() { 
  list(a=eps,b=rep(eps,K))
}
parameters <- c("a","b")

result.sim <- bugs(data2, inits, parameters, 
                   model.file="brand2.bugs",
                   n.chains = 4, n.iter = 1000,
                   debug=F,
                   working.directory=getwd())

result.sim$sims.list$beta1
print(result.sim, digits=3)

plot(as.mcmc(result.sim$sims.matrix))