--- On Thu, 11/2/10, Kate Perper wrote:
> I am working with multinomial logit models and
> using survey data. I would like to see if adding
> interactions to my model improves the fit. Given
> that I'm working with survey data, what is the best
> way to compare model fit?
You use -test- on the interaction terms, see: <http://www.stata.com/support/faqs/stat/lrtest.html>.
Testing whether this interaction term equals zero is not
to everybody's taste (in particular those who look
at marginal effects), but if interpret your effects in
terms of odds ratios/relative risk ratios) then it
does make sense. <http://www.maartenbuis.nl/wp/
interactions.html>
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/
