model {
for (i in 1:N) {
for (j in 1:3) {
y[i,j] ~ dnorm(psi[i,j], inv.sigma.squared)
psi[i,j] <- alpha[i] + beta[i]*(t[i,j] - tbar)
+ gamma*(y0[i] - y0bar)
}
alpha[i] ~ dnorm(mu.alpha, inv.omega.alpha.squared)
beta[i] ~ dnorm(mu.beta, inv.omega.beta.squared)
}
inv.sigma.squared <- 1/sigma.squared
inv.omega.alpha.squared <- 1/omega.alpha.squared
inv.omega.beta.squared <- 1/omega.beta.squared
sigma.squared <- pow(sigma, 2)
omega.alpha.squared <- pow(omega.alpha, 2)
omega.beta.squared <- pow(omega.beta, 2)
log(sigma) <- log.sigma
log.sigma ~ dunif(-10, 10)
omega.alpha ~ dunif(0, 100)
omega.beta ~ dunif(0, 100)
mu.alpha ~ dnorm(0, 0.0001)
mu.beta ~ dnorm(0, 0.0001)
gamma ~ dnorm(0, 0.0001)
y0bar <- mean(y0[])
}
Inits:
list(log.sigma = 0, omega.alpha = 0.1, omega.beta = 0.1,
mu.alpha = 0, mu.beta = 0, gamma = 0)
Data...
Dbar = post.mean of -2logL; Dhat = -2LogL at post.mean of stochastic nodes
Dbar Dhat pD DIC
y 813.563 715.127 98.436 911.998
total 813.563 715.127 98.436 911.998
node mean sd MC error 2.5% median 97.5% start sample
deviance 813.6 22.05 0.3335 769.1 813.7 856.8 1001 100000
gamma 0.6711 0.08489 9.3E-4 0.5043 0.6712 0.8374 1001 100000
mu.alpha 6.137 0.1504 5.448E-4 5.841 6.136 6.433 1001 100000
mu.beta -1.075 0.1403 0.003874 -1.343 -1.075 -0.7964 1001 100000
omega.alpha 1.428 0.1197 5.379E-4 1.209 1.422 1.68 1001 100000
omega.alpha.squared 2.053 0.3467 0.001557 1.462 2.022 2.822 1001 100000
omega.beta 0.3042 0.2114 0.00826 0.01907 0.2695 0.7795 1001 100000
omega.beta.squared 0.1373 0.1693 0.005531 3.635E-4 0.07264 0.6076 1001 100000
sigma 0.9951 0.05628 5.927E-4 0.8895 0.9933 1.111 1001 100000
sigma.squared 0.9933 0.1127 0.001173 0.7911 0.9866 1.234 1001 100000