Hi all, I am conducting a series of multinomial logistic regression using complex survey data (I used "svy: mlogit" for this). I was wondering if there is any way I can conduct a test of model fit improvement between two nested models using survey data(i.e., [svy: mlogit y x1 x2] compared to [svy: mlogit y x1 x2 x3] like a likelihood ratio test in usual mlogit models. Would a Wald test in post estimation do the job? Any suggestions would be greatly appreciated. Thanks.
Jinseok
*
* 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/
