Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Marco Francesco" <cariboupad@gmx.fr> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: bug in margins |
Date | Thu, 24 Jan 2013 11:38:30 +0100 |
Dear all, I have small problem. Before the 18dec update (Stata 12 MP) I ran the following regression -reg Y c.X##(T P T#P), vce(cluster group)- That is a continuous variable X is fully interacted with the variables T, P, and the interacted T#P. Think of if as a main regression (T P T#P) that I want to further specify by allowing interaction terms with another continuous regressor X. Unfortunately after the the 18dec update I ran exactly the same regression (same data, same everything, etc) and I got completely different results... My question is : is this related to the fix included in the 18dec update : " margins no longer allows interaction terms to contain continuous variables in the margins list." Should I trust the results now ? Many thanks for your help * * 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/