Dear Statalisters,
I want to estimate the impact of an invariant variable in a panel data framework. Since the Hausman test rejected the use of the random effect model, I have to use the Hausman-Taylor estimator.
However, using the xthtaylor command that performs Hausman and Taylor estimation in stata9, stata "asks" me to specify not only time-varying but also time-invariant endogenous regressors, while I have only time-varying endogenous regressors.
Including no time-invariant endogenous regressors implies an error message...including some time-invariant endogenous regressors (almost randomly) results in strange results (unsurprisingly).
Does anyone have any idea to solve this problem?
Many thanks
Chris
*
* 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/
