R Package about Variable Selection for GLMM (Generalized Linear Mixed Model)?

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

R Package about Variable Selection for GLMM (Generalized Linear Mixed Model)?

Jun Bum Kwon


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]]


______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|

Re: R Package about Variable Selection for GLMM (Generalized Linear Mixed Model)?

bbolker
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

______________________________________________
[hidden email] 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.