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: Marginal Effects for Logistic Mulitlevel Model withInteractionTerms


From   Guy Grossman <guygrossman1@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Marginal Effects for Logistic Mulitlevel Model withInteractionTerms
Date   Fri, 20 Aug 2010 07:18:54 +0300

Dear Iris -

I recommend you check out the article  "Understanding Interaction
Models: Improving Empirical Analyses" (Political Analysis 14: 63-82)
and the accompanying http://homepages.nyu.edu/~mrg217/interaction.html

The authors provide Stata code for calculating marginal effects after
logistic regression, which you can easily adopt to the xtmelogit.

Cheers,
Guy


On Fri, Aug 20, 2010 at 12:18 AM, <Iris.C.Goensch@zeu.uni-giessen.de> wrote:
>
> Thanks a lot for all your help!
> Unfortunately I'm still not sure what to do:
>
> -mfx does not recognize interaction terms and thus does not calculate the marginal effects for interaction terms in the correct way (the problems resulting are explained in Ai/ Norton and Greene (2010).
>
> -Does it make sense not to calculate any marginal effects but just to include interaction terms in the multilevel logit model and estimate that with the stmelogit command and the odds ratio (or) option? Or do I have the same conceptual problem that I can't interpret odds ratios for interaction terms?
>
> Best,
> Iris
>
> Zitat von Maarten buis <maartenbuis@yahoo.co.uk>:
>
>> --- On Thu, 19/8/10, Iris.C.Goensch@zeu.uni-giessen.de wrote:
>>>
>>> I am estimating a logistic multilevel model (which
>>> includes interaction terms) and I would like to calculate marginal
>>> effects, but I always get the error "default prediction is a
>>> function of possibly stochastic quantities other than e(b)".
>>
>> The marginal effect depends on all covariates including the random
>> effects. The random effects is the stochastic quantity that the
>> error message complaints about. I think that the easiest solution
>> is to report odds ratios. This has the added advantage that it
>> also makes the interpretation of the interaction effects much
>> easier.
>>
>> Hope this helps,
>> Maarten
>>
>> --------------------------
>> Maarten L. Buis
>> Institut fuer Soziologie
>> Universitaet Tuebingen
>> Wilhelmstrasse 36
>> 72074 Tuebingen
>> Germany
>>
>> http://www.maartenbuis.nl
>> --------------------------
>>
>>
>>
>>
>> *
>> *   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/
>>
>
>
>
> --
>
> Iris Gönsch
> Research Assistant
> Centre for International Development and Environmental Research (ZEU)
> University of Giessen
> Senckenbergstr. 3
> D-35390 Giessen, Germany
>
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
>
> *
> *   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/



--
Guy Grossman
Department of Political Science
Columbia University

*
*   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