Hello all,
I tried to run a multilevel mixture Rasch model using OpenBUGS. When I generated initial values, I received an error message "Something went wrong in procedure Node.value in module GraphMixture." Since I have limited knowledge about programming, I googled and tried enabling "Oberonmicrosystems" trap handler. The only information I got from the trap handler is something went wrong at theta[2876]. But, I am not able to find where the code is wrong and how to handle it. I tried to change priors and parameters, and also tried to upload some initial values. The model still didn't work. I attached the model below. If you need data or anything, please let me know.
Any comments are welcome. I appreciate your time. Thanks a lot!
model {
for (i in 1:3000) {
for (j in 1:40) {
y[i,j] ~ dbern(eta[i,j])
logit(eta[i,j]) <- psi[s[i],t[tID[i]]]*theta[i]-b[j,s[i],t[tID[i]]]
}
theta[i] ~ dnorm(mu[s[i],t[tID[i]]],tau0[s[i],t[tID[i]]])
}
mu[1,1] <- 0
mu[2,1] ~dnorm(0,1)
mu[1,2] ~dnorm(0,1)
mu[2,2] ~dnorm(0,1)
tau0[1,1] <- 1
tau0[1,2] <- 1
tau0[2,1] <- 1
tau0[2,2] <- 1
for (i in 1:2) {
for (j in 1:3) {
psi[i,j] ~ dgamma(1,1)
}
}
#student level
for (i in 1:3000) {
s[i] ~ dcat(pi[t[tID[i]],1:2])
}
for (i in 1:3) {
for (j in 1:2) {
pi[i,j] <- delta1[i,j]/sum(delta1[i,])
delta1[i,j] ~ dgamma(alpha1[j],1)
}
}
#school level
for (k in 1:100) {
t[k] ~ dcat(pi1[1:3])
}
for (i in 1:3) {
pi1[i] <- delta[i]/sum(delta[1:3])
delta[i] ~ dgamma(alpha[i],1)
}
# pi1[1] <- 0.85
# pi1[2] <- 0.09
# pi1[3] <- 0.06
for (i in 1:100) {
for (j in 1:2) {
for (k in 1:3) {
b[i,j,k] ~ dnorm(0,1)
}
}
}
}
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|