Yesterday I posted a question regarding interpretation of an error message
I was getting from WinBUGS.
Summary of responses:
1) I had accidentally forgotten to treat the normal priors as having
precisions, the priors I wrote were variances; this led to:
2) Putting my model in a portion of the parameter space that was
numerically unstable given the functional form.
One commentator noted that I was fitting a lot of parameters relative to
the amount of data. That was true (the model as specified has many
parameters), and contributed to the problem.
THANKS to Bill Bell, David Speigelhalter, and Margaret Short, who each
supplied important clues for solving this.
Dean H. Judson, Ph.D., Mathematical Statistician and Group Leader
Administrative Records Evaluation and Linkage Group
Planning, Research and Evaluation Division
U.S. Bureau of the Census
Washington, DC 20233
Phone: 301-457-4222
Fax: 301-457-3066
email: [log in to unmask]
-------------------------------------------------------------------
To mail the BUGS list, mail to [log in to unmask]
You can search old messages at www.jiscmail.ac.uk/lists/bugs.html
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
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]
|