Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Manos Mentzakis <mmentzakis@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: margins atmeans for discrete changes |
Date | Mon, 4 Feb 2013 18:03:02 +0000 |
Dear all, I have a 3-level categorical variable and depending on whether it is specified as -i.levels- or -i.level2 i.level3- margins gives slightly different values. My manual calculation and mfx agree with the second specification. Could anyone point out what exactly is the difference in the marginal effect calculation? Thank you Regards Manos ** EXAMPLE sysuse nlsw88.dta, clear drop if occupation>3 | occupation==. | union==. tab occupation,g(oc) * probit union i.occupation margins, dydx(*) atmeans * probit union i.oc2 i.oc3 margins, dydx(*) atmeans * probit union oc2 oc3 mfx * * 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/