Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Sriram Narayanan <ssriram74@yahoo.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | st: model selection tests in dynamic panel |
| Date | Mon, 1 Nov 2010 13:07:31 -0700 (PDT) |
I am trying to estimate a model using the xtdpdsys command. I would ideally like
to do some nested model comparisons since I have some hypothesis revolving
around the interaction effects. I guess doing the LR test in this case may not
be possible since there is no log-likelihood number. However, I was wondering if
it is reasonable to do a chi-square difference test in this context to compare
nested models? If some of the group members have insight on this, I would be
very grateful.
Regards
Sriram
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/