Print

Print


Hi BUGSians!
I've got a problem with sampling variance components in a mixed model.
There is no problem with compiling or updating. But when I try to sample
(Inference...Samples) the variancecomponents (sigma2.fehler,
sigma2.gold, sigma2.interaktion), WinBUGS gives me following error -
message: "monitor could not be set". It is very interesting, that there
is no problem to sample nodes like "alpha" or something else. Whats
wrong, any ideas? Is it because of the large number of missing values in
the dataset? SAS is able to estimate, of course that are nice ones. I
will compare it with WinBUGS-estimations for my dissertation, help me,
please. Thx.
Marc

Thats the model:

model
{
     for(j in 1:b)
    {
        beta[j]~dnorm(1.0E-10,1.0E-10)
    }
    for(i in 1:a)
    {
        alpha[i]~dnorm(alpha0,gold)
        for(j in 1:b)
        {
            interaction[i,j]~dnorm(interaction0,interaktion)
            mu[i,j]<-mw+alpha[i]+beta[j]+interaction[i,j]
            for(k in 1:nij)
            {
                y[(i-1)*6+(j-1)*2+k]~dnorm(mu[i,j],fehler)
            }
        }
    }
    for(j in 1:b)
    {
            SAS_Estimates_Herst[j]<-mean(mu[,j])
    }
    mwdiff[1]<-SAS_Estimates_Herst[1]-SAS_Estimates_Herst[2]
    mwdiff[2]<-SAS_Estimates_Herst[1]-SAS_Estimates_Herst[3]
    mwdiff[3]<-SAS_Estimates_Herst[2]-SAS_Estimates_Herst[3]
    alpha0~dnorm(1.0E-10,1.0E-10)
    interaction0~dnorm(1.0E-10,1.0E-10)
    mw~dnorm(1.0E-10,1.0E-10)
    sigma2.fehler<-1/fehler
    fehler~dgamma(0.001,0.001)
    sigma2.gold<-1/gold
    gold~dgamma(0.001,0.001)
    sigma2.interaktion<-1/interaktion
    interaktion~dgamma(0.001,0.001)
}
#daten
list(a=8,b=3,nij=2,y=c(NA,NA,NA,NA,NA,NA,NA,NA,NA,NA,803,NA,NA,715,NA,772,NA,NA,792,NA,698,NA,620,NA,NA,NA,NA,743,NA,NA,907,NA,NA,NA,NA,NA,NA,NA,NA,792,NA,734,835,NA,870,NA,NA,681)
#initial's
list(mw=700,fehler=0.0001,gold=0.0001,interaktion=0.0001)

-------------------------------------------------------------------
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