Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Fred Dzanku <fdzanku@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Heteroskedastic Probit Estimation |
Date | Sat, 6 Nov 2010 09:54:59 +0000 |
Dear All, I am estimating a heteroskedastic probit model but I am not too sure which variables to include in the heteroskedastic function. This is because a variable addition test shows that not including as explanatory variables the variables I specify in het(*) causes my model suffers from omitted variables bias. What do I do given Cameron and Trivedi (Page 455)'s advice not to include the variables in het(*) also as explanatory variables? I am a bit confused about all these. Can someone please help? Relatedly, is there any way of explicitly testing which variables to include in het(*)? Many thanks. Fred * * 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/