Re: st: Marginal Effects for Logistic Mulitlevel Model withInteractionTerms

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Mon, 30/8/10, Richard Williams wrote:
> I agree that we want some quantification of the size of
> interaction effects. But, is the marginal effect of the
> interaction term no longer even a candidate for that
> task, since margins does not report it? 

The one reported by -mfx- never was a proper candidate for
that. The marginal effects reported by -inteff- and 
-inteff3- (see: -findit inteff- and -findit inteff3-) are
still proper candidates. I am hopeful that -margins- will at
some point in the future report meaningfull marginal
effects for interaction terms. However, I have no basis 
for that hope other than that with the new factor variable
notation they have all information they need to compute 
those.

--- 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/