My model is conditional logit with alternative-specific constants (i.e., alternative dummies) based on McFadden's choice model.
clogit y a1-a16 x1 x2 x3, group(id)
I performed the above model for two different samples and want to compare the difference of coefficients between two samples.
I made the interaction terms of all independent variables(a1-a16, x1, x2, x3) with group dummy(t) and run the test for pooled sample like following.
clogit y a1-a16 a1Xt-a16Xt x1 x2 x3 x1Xt x2Xt x3Xt, group(id)
Is this a right way? I am not sure if this well-known method is also appropriate to conditional logit model.
Furthermore, the above model does not work (STATA is shut down).
I guess that this problem is probably caused by the fact that I included alternative-specific constants.
Please, help me, someone.....
Yuri Jo
*
* 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/
