Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Nick Cox <n.j.cox@durham.ac.uk> |
| To | "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |
| Subject | RE: st: Marginal Effects LogNormal Two-Part Model |
| Date | Mon, 21 Mar 2011 14:21:11 +0000 |
This is the third time in as many days that this has been posted. The answer appears to be Sorry, but no; and also do please pay attention to protocol on repeated posts at http://www.stata.com/support/faqs/res/statalist.html#noanswer -- which you were asked to read when you joined Statalist. Nick n.j.cox@durham.ac.uk Clifton Chow I am wondering if someone could point me to the code or user defined program in STATA for calculating marginal effects (especially unconditional ME) for a Two Part Probit/LogNormal model (alternative to Tobit). My program is below: First Part (Probit) P(HoursWork | $xlist) probit binarywork age female white educ faminc healthstat Second Part (Log Normal) E(LNHours | $xlist, hourswork >0) regress LNHours age female white educ faminc healthstat if binarywork==1 I have gone through the statalist archives and didn't come across any for logNormal two-part models. Any help is greatly appreciated. * * 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/