Hi,
I use stata version 6. I am trying to test the
validity of an mlogit model. While doing Hausman IIA
test, I encounter some problems like having negative
chi-square results for some of the categories. And for
some categories, it does not even give a result, but
give an error message of "big is not found r(111)". I
guess the negative chi-squared results are related to
asymptotic assumptions of the test. But I can not
figure out the latter problem. Any ideas, why it can
be?
I see that there are two alternatives for doing IIA
test. One is doing the hausman test with suest. The
second one is smhsiao function.
I found ado files of smhsiao on the web. It worked for
giving the mlogit results, but did not give the test
results. The error message is "eq() invalid r(198)".
Would this error be related to the version I am using?
Suest alternative on the other hand again does not
exist in my version. Is there a module or a program
that I can download from the web (as the one for
smhsiao) to do this test?
Regards,
Uygar
__________________________________
Do you Yahoo!?
Dress up your holiday email, Hollywood style. Learn more.
http://celebrity.mail.yahoo.com
*
* 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/
