Statalist


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

st: Intercept estimates in -nlogit- with case-specific variables


From   Mike Lacy <Michael.Lacy@colostate.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Intercept estimates in -nlogit- with case-specific variables
Date   Tue, 18 Mar 2008 16:10:00 -0600

I'm trying to run a nested logit model on data with all case-specific predictors, and I am confused about the presence/role of the intercept. What confuses me is that none of the examples in the manual entry on -nlogit- show any estimate of the intercept, although I have seen this elsewhere.

My understanding that with case-specfic predictors, the intercept estimate is meaningful, and can be obtained with a proper data setup, by which I mean that for each case, there are records in the file for each alternative, with indicator variables to mark each possible alternative, and indicator X covariate product variables for each case-specific covariate. This kind of setup, if used with -clogit- with case-specific predictors, produces intercept (and other) estimates identical to -mlogit-.

So, is it appropriate to use these indicator terms to to pick up the intercept estimates in an -nlogit- that has only case--specific variables?

Regards,
=-=-=-=-=-=-=-=-=-=-=-=-=
Mike Lacy, Assoc. Prof.
Soc. Dept., Colo. State. Univ.
Fort Collins CO 80523 USA
(970)-491-6721

*
* 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