R help - Dec 2008 - WinBUGS posterior samples (via R2WinBUGS)?

Hi all,

I did some analysis using package R2WinBUGS to call WinBUGS. I set the
iterations to 50000 (fairly a large number, I think), but after the program
was done,  the effective posterior samples contained only 7 draws.  I don't
know why.

By the way, I checked posterior sample size by using bugsobj$n.sims. And,
for my previous practice with WinBUGS/R2WinBUGS, no such strange thing
happend. Does anybody have some idea on what might be wrong with my model or
whatever part in this program?  Appreciate it.

Anny

	[[alternative HTML version deleted]]

3 comments:

1. Quoting the last lines of this message:
"PLEASE do read the posting guide 
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code."
Otherwise we cannot help anyway.

2. This is not the BUGS mailing list.

3. I do prefer BRugs, the R package that is based on OpenBUGS, because 
it is closer to BUGS, more flexible, sometimes easier to debug, ....

Best wishes,
Uwe Ligges




Anny Huang wrote:
> Hi all,
> 
> I did some analysis using package R2WinBUGS to call WinBUGS. I set the
> iterations to 50000 (fairly a large number, I think), but after the program
> was done,  the effective posterior samples contained only 7 draws.  I
don't
> know why.
> 
> By the way, I checked posterior sample size by using bugsobj$n.sims. And,
> for my previous practice with WinBUGS/R2WinBUGS, no such strange thing
> happend. Does anybody have some idea on what might be wrong with my model
or
> whatever part in this program?  Appreciate it.
> 
> Anny
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

Uwe Ligges <ligges <at> statistik.tu-dortmund.de> writes:
> 3. I do prefer BRugs, the R package that is based on OpenBUGS, because 
> it is closer to BUGS, more flexible, sometimes easier to debug, ....
Which I strongly second, not only because of the author.

Dieter

Anny Huang <annylhuang <at> gmail.com>
writes:
> I did some analysis using package R2WinBUGS to call WinBUGS. I set the
> iterations to 50000 (fairly a large number, I think), but after the program
> was done,  the effective posterior samples contained only 7 draws.  I
don't
> know why.
This indicates that you have a mixing problem! I guess that you do not have 
enough information to retrieve the post. dist. of that parameter. You might
try to reconsider your model or reparametrize it.

gg