Femke Lambers wrote:
>>I hope someone can help me with the following:
I'm fitting a multivariate model with GEE (fam (binomial) link(logit)
robust eform). My output shows me the p-value of each category of each
variable. Does anyone know how I can get the overall p-value of each
variable (on its own, so not per category)? I would also like to know
how I test the goodness of fit in this kind of model. Hopefully, you can
help me out. Thank you!<<
Things may have changed but from what I recall there isn't a global GOF
test for GEE and you have to use Wald tests to do inference. You may
want to see Hardin and Hilbe (2003):
http://www.stata.com/bookstore/gee.html. The recent book by Molenberghs
and Verbeke (Models for Discrete Longitudinal Data, 2005, Springer)
should also discuss this problem.
Jay
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/