st: RE: model fit comparison for xtgee

["Nick Cox" <n.j.cox@durham.ac.uk>]

This is an FAQ: 

How can I get an R-squared value when a Stata command does not supply one? 

http://www.stata.com/support/faqs/stat/rsquared.html

Nick 
n.j.cox@durham.ac.uk 

tlum@umn.edu

I am comparing two xtgee models to see how much explanatory power has 
increased
by adding a block of variables. I did a hausman test and the null 
hypothesis
was rejected. I do want to know how much more variance was explained by the 
new variables. Can the xtgee generate something like r-square type model 
fit index. Or is there anyway I can come up with some formal indicators for 
improvement in model fit under the new model.


*
*   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/