Dear Statalist,
I am trying to run a Heckman selection model. I tried both the maximum-
likelihood method and the two-step estimator and they produced very
different results, moreover the rho=1 with the two-step method. I
suspected that it might be because my data did not support joint
normality. So I estimated again the model including the square and the
cube of some of my variables and tested the joint significance of these
variables which was accepted. I am a bit blocked now, is there any way
to verify the joint normality? And if it is rejected is there any method to
overcome this problem?
Thanks for your help,
Miriam
*
* 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/
