Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Cameron McLoughlin <cameron.mcloughlin@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Mixlogit Error |
Date | Fri, 2 Aug 2013 10:06:20 +0200 |
Dear Statlisters, I have a question regarding the correct use of the user written command mixlogit. I have a monthly panel of 27 countries over 7 years and am trying to estimate a mixed logit model with a crisis dummy (0/1) as the dependent variable. My data is stacked in the usual STATA format, for example: Id Time Cisis Var1 Var2 1 1994m1 0 1 1994m2 1 1 1994m3 0 Following previous advice on this forum available in other FAQs regarding mixed logit, I first made sure that I can estimate a clogit model using this data. I have no problems using for example with the command: clogit crisis var1 var2, group(id) vce(cluster id) After reading Arne Hole's STATA Journal article on the mixed logit command, I thought that the way I should code this command should be: mixlogit crisis, rand(var1 var2) group(time) id(id) When I try this piece of code though, I receive the error message 'at least one group has more than one chosen alternative'. Any ideas as to what I am doing wrong? I guess that I am not specifying the group() option i.e. the choice sets correctly, but I'm still strugglin with how to specify this option..... Thanks alot for any suggestions you may have!!! Cameron McLoughlin * * 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/