Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Igor Ramzak <igor.ramzak@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: margins after conditional logit |
Date | Fri, 14 Jun 2013 17:29:03 +0200 |
Dear Statalisters, I am running the -clogit- command in order to explain the probability on cardiovascular disease. Two of my explanatory variables are dummies that are mutually exclusive, sush as age groups and race. These are included in the estimation using "i.age" and "i.race", so the full model is: -clogit cv_disease i.age i.race weight weight_squared work- When I run: -margins, predict(pu0) dydx(*)- the table shows for these age and race dummies "(not estimable)". What is the reason for this error? Where is my mistake? Is there an alternative way using Stata? Or is it because of the way of estimating (using clogit)? Best, Igor * * 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/