Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Backward Selection (AIC) with xtmixed


From   SR Millis <aa3379@wayne.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Backward Selection (AIC) with xtmixed
Date   Sat, 16 Jul 2011 12:08:51 -0700 (PDT)

The problems with the various flavors of stepwise variable selection (including forward and backward selection) have been discussed quite exhaustively on Statalist (search the archives).  In short, the models/variables selected by stepwise methods tend to be unstable and often do not replicate well.  

Scott Millis



--- On Fri, 7/15/11, Allan Kennedy <dkstata@gmail.com> wrote:

> From: Allan Kennedy <dkstata@gmail.com>
> Subject: Re: st: Backward Selection (AIC) with xtmixed
> To: statalist@hsphsun2.harvard.edu
> Date: Friday, July 15, 2011, 10:53 PM
> Could you provide a deeper rationale
> as to why this approach is not
> recommended?/
> 
> On Fri, Jul 15, 2011 at 10:44 PM, SR Millis <aa3379@wayne.edu>
> wrote:
> > This procedure is not recommended.
> >
> >
> > ~~~~~~~~~~~
> > Scott R Millis, PhD, ABPP, CStat, PStat®
> > Professor
> > Wayne State University School of Medicine
> > Email:  aa3379@wayne.edu
> > Email:  srmillis@yahoo.com
> > Tel: 313-993-8085
> >
> >
> > --- On Fri, 7/15/11, Allan Kennedy <dkstata@gmail.com>
> wrote:
> >
> >> From: Allan Kennedy <dkstata@gmail.com>
> >> Subject: st: Backward Selection (AIC) with
> xtmixed
> >> To: statalist@hsphsun2.harvard.edu
> >> Date: Friday, July 15, 2011, 10:36 PM
> >> Greetings,
> >>
> >> I am interesting is using an automated command
> that will
> >> estimate
> >> model fit with a backward selection model with AIC
> of an
> >> --xtmixed--
> >> approach. I know that both --stepwise-- and
> --swaic-- both
> >> do not
> >> support xtmixed but I wanted to see if there were
> any other
> >> user
> >> written programs that support --xtmixed--. An
> example of my
> >> code is as
> >> follows:
> >>
> >> xtmixed OUTCOME  10PREDICTORS  CONTROL
> VARIABLES
> >> ||RANDOM_1:
> >> ||RANDOM_2: ||RANDOM_3: , covariance(unstructured)
> var mle
> >>
> >> I would like to estimate the lowest AIC model of
> any
> >> combination of
> >> the 10 predictors.
> >>
> >> Your assistance would be greatly appreciated.
> >>
> >>
> >> --AK
> >> *
> >> *   For searches and help try:
> >> *   http://www.stata.com/help.cgi?search
> >> *   http://www.stata.com/support/statalist/faq
> >> *   http://www.ats.ucla.edu/stat/stata/
> >>
> >
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> >
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index