Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: gee p-value and goodnesss of fit


From   "Verkuilen, Jay" <JVerkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: gee p-value and goodnesss of fit
Date   Wed, 12 Mar 2008 11:06:58 -0400

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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index