Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Grethe Søndergaard <dkstatstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: overall test of interaction |
Date | Mon, 15 Nov 2010 13:34:15 +0100 |
Dear list, I am analysing my data using a stratyfied cox-regression analysis Since I want to examine if any interaction occurs between education and birthyear, I have inserted an interaction-term: stcox i.education i.birthyear i.education#i.birthyear, strata(group-id) vce(cluster person-id) Is it possible for me to do an overall test in order to see, if the interaction is significant? Since I am using vce(cluster), I cannot do a lrtest of a model with and without the interaction term... Best regards Grethe * * 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/