I have had two recent problems and wonder if anyone can advise on the
possible causes/remedies:
1) I can't get classic BUGS to run at all on an IBM ThinkPad 600 running
Win 95. The method I have used on several computers using a DOS window
just doesn't seem to run (BUGS crashes immediately and hangs the DOS
window) -- does anyone know any setting I need to change to make the IBM happy?
2) The following code:
{
for( i in 1 : Num ) {
y0[i] ~ dbin(p0[i], n0[i]);
y1[i] ~ dbin(p1[i], n1[i]);
logit(p0[i]) <- beta[i,1] - beta[i,2]/2;
logit(p1[i]) <- beta[i,1] + beta[i,2]/2;
beta[i,] ~ dmnorm(mu[], tau[,]);
}
mu[] ~ dmnorm(mean[],prec[,]);
tau[,] ~ dwish(R[,], 2);
...
(to run a Bayesian meta-analysis) works fine with the older BUGS version
0.6 (Dec 96) but with 0.603 (Mar 98) it crashes on the first update with
this message:
Bugs>update(1)
time for **** gp.rts: real compare with NaN in module DFreeARS at line 43
****
----------------------------------------------------------------------
John Carlin <[log in to unmask]>
Dept of Epidemiology & Biostatistics
College of Public Health, MDC-56
University of South Florida
13201 Bruce B. Downs Blvd
Tampa, Florida 33612
Tel. 1-813-974-6673 (office) 974-4860 (messages); 1-941-342-6533 (home)
Fax. 1-813-974-4719
(on sabbatical Jan-Dec 1999; otherwise:
Clinical Epidemiology & Biostatistics Unit,
Royal Children's Hospital Research Institute & University of Melbourne,
Melbourne VIC 3052, AUSTRALIA)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|