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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Beta's and xtmixed


From   "JVerkuilen (Gmail)" <jvverkuilen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Beta's and xtmixed
Date   Mon, 22 Oct 2012 10:31:43 -0400

On Mon, Oct 22, 2012 at 5:56 AM, Filip Van Droogenbroeck
<filip.van.droogenbroeck@vub.ac.be> wrote:
> Hello, everybody,
>
> I'm trying to get beta's for the -xtmixed- multilevel regression
> command. There is no beta option for the xtmixed command. Anybody knows if there is a good reason for this? By using esttab I can get beta's after xtmixed. Can these beta's be trusted after the xtmixed command? Or are they only correct when doing normal linear regression?

Standardized regression coefficients, which is what I assume you want,
don't really mean the same thing in a mixed logit model as in linear
regression, because while the variance of the x variables is
meaningful, the variance of the linear predictor is set by choice of
the logit link to pi^2/3. You really can't compare them across models
readily. Usually you want to use standardized regression coefficients
across models, so this is a problem.

See Bauer, D.J. (2009). A note on comparing the estimates of models
for cluster-correlated or longitudinal data with binary or ordinal
outcomes. Psychometrika, 74, 97-105.
http://www.unc.edu/~dbauer/manuscripts/Bauer_2009.pdf

Chances are good you can get what you want with predicted
probabilities computed on relevant scenarios, though that's not always
true.

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


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