Dear Sirs,
may be somebody could help me with the following problem.
I would like to use "Stochastic Search Variable Selection" method introduced by
George and McCulloch, that is I have parameters p[j] that must be dependent on
the state of the indicator variable g[j]. If g[j]=0, the parameters must be near
0, if g[j]=1, they are near 1.
I try to describe priors on the parameters. The problem is that the parameters
are in (0,1), so I use the logistic transformation and some other tricks to make
the parameter priors dependent on g[j]. Please, look into the following:
p[j] <- exp(l[j])/(1+exp(l[j]));
l[j] <- pow(-1,1-g[j])*k[j];
k[j] ~ dlnorm(m,t);
m<-0.0;
t<-0.01;
At this specification I get a bug:
"real compare with NaN in module determ at line 384".
Could anybody say, what is the problem and whether my prior specification has
sence.
Thank you very much,
Svetlana Bulashevska.
-------------------------------------------------------------------
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
|