R help - Aug 2010 - R Package about Variable Selection for GLMM (Generalized Linear Mixed Model)?

Hi all, 



I have searched for a long time to find out R program about V ariable S election
for GLMM (Generalized Linear Mixed Model).



I saw several great R packages for V ariable S election. I  also found several R
packages for GLMM. But, I did not find yet R package about V ariable S election
for GLMM even though sevel  papers about it have been published.  



In fact,  I need V ariable  S election tool for Aggregate RC (Random
Coefficient) Logit Model. However, I don't think that Aggregate RC Logit is
popular since most researcher use subject (individual) level choice data rather
than aggregate data which only has information about the summation across all
individual in the given time period.



Therefore, if I could find out any Variable Selection tool for RC logit (one
of GLMM), I would like to apply it for my problem, which is "Variable
Selection for Aggregate RC Logit.





Sincerely, 

Jun-Bum
	[[alternative HTML version deleted]]

Jun Bum Kwon <junbum <at> uwm.edu> writes:

[lightly edited]
> I saw several great R packages for Variable Selection. 
> I also found several R packages for GLMM. But, I
> did not find yet R package about Variable Selection for 
> GLMM even though several? papers about it have
> been published.
  Such methods may or may have been coded in R at present.  
It would be helpful to (a) post references for the 'several papers' 
you refer to and (b) post further queries to the R-sig-mixed-models
mailing list, which is more appropriate for questions about (G)LMMs.

  This may also be an appropriate place for the comment, "tell us what
problem you are trying to solve, not just which methods you want to use",
since
sometimes people on the list are aware of different (and/or better, at
least according to some criteria) methods for solving the same problem.

  cheers
    Ben Bolker