Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: To: statalist@hsphsun2.harvard.edu


From   "Suzanne Bouffard" <bouffasu@gse.harvard.edu>
To   undisclosed-recipients: ;
Subject   st: To: statalist@hsphsun2.harvard.edu
Date   Sun, 22 May 2005 11:36:13 -0400

I am using svymlogit to perform a multinomial logistic regression with survey data. A couple of questions:

-Is there a way to assign a unit to a continuous predictor? For example, in SAS, I could assign units of 10,000 to my income variable, so that the coefficient and odds ratio will correspond to increases of $10,000 rather than increases of $1, which is not very meaningful.

-svymlogit gives me an overall F value with df based on strata and psu's. Is this F value the same as -2LL? If not, is there some way I can get that value?

- Is there a way to test the whole set of possible interactions - for example, in a nested fashion - before including each individual interaction? And should I be using inteff? Why can't I just enter the interaction terms into the svymlogit model? What is the logic behind inteff?

thanks.

*
* 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