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

st: Hausman test for IIA, anyone??


From   "Robert Bozick" <rbozick@smtp.aed.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Hausman test for IIA, anyone??
Date   Thu, 06 Nov 2003 09:21:19 -0500

Hey Stata users -- I posted a question a few days back and haven't heard from anyone, (which means either no one knows the answer or the message may have gotten buried in your inbox).  In any case, I am reposting it again --- any help/thoughts are appreciated!

I am having a problem using the Hausman test of independence of irrelevant alternatives (IIA) following the estimation of a multinomial logistic regression model.  I have a three category nominal dependent variable called DV (0 = no college, 1 = immediate enrollment in college, and 2 = on-time enrollment in college).

I first estimate the model:
mlogit DV x1 x2 x3 x4

STATA estimates the model with no problem.  I then use the commands to obtain the Hausman test:

mlogtest, hausman base 

STATA tells me
**** Hausman tests of IIA assumption

Ho: Odds(Outcome-J vs Outcome-K) are independent of other alternatives.

no coefficients in common; specify -equations(matchlist)- for the problem of different equation names. r(498);


Does anyone know if I am doing something wrong?

Thanks in advance for your help,
Robert


*
*   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