st: Marginal Effects from Tobit + FE

[mehmet can <micdepo@gmail.com>]

Hello Statalisters,


This email is with regards to computing marginal effects from Tobit
regressions with fixed effects (please refer to Greene (2004) for the
justification for using FE with Tobit). When I compute marginal
effects (i.e. margins, dydx(*) atmeans) after my Tobit regression, I
see that Stata rightfully takes into account the means of the fixed
effect variables, too. I couldn't make sure if this is something I
should fix or neglect, so I spent some time looking at earlier
discussions on Statalist. Nonetheless, I couldn't really make sure how
I can solve this problem in Tobit framework or whether it is actually
a critical issue. I would very much appreciate any insights. Thanks
very much in advance.


All the best,
Mehmet
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/