Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Sezgin Ayabakan <ayabakan@utdallas.edu> |
To | "." <statalist@hsphsun2.harvard.edu> |
Subject | st: Murphy-Topel correction for MLogit - Tobit is possible? |
Date | Thu, 22 Aug 2013 11:42:36 -0500 (CDT) |
Dear Stata community, I have been using Murphy-Topel correction for a logit (first eq.) - tobit (second eq.) model and I've referred to the paper by Hole (2006), Stata Journal 6 (4). However, now I have a categorical variable (3 different cases), so in order to handle this situation I included two binary variables in the tobit model. And these two binary variables are determined within in the system (through exogenous variables and some other variables-instruments). At this point, I'm not sure if I can apply Murphy Topel correction approach. I thought of using mlogit in the first equation, and then getting the predicted probabilities for cases 2 and 3, and using these predicted probabilities in the tobit model. Would mlogit-tobit be possible? This is my first posting, so hopefully everything is clear. Thank you, -- Sezgin Ayabakan * * 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/