Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Wiemann, Markus" <m.wiemann@fs.de> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: nlogit error because of replicate levels |
Date | Thu, 26 Aug 2010 15:39:06 +0200 |
Hello, Using the nlogit command (in Stata/SE 10.0) I repeatedly run into an error I do not find the source of. I have the following tree structure . nlogittree trigger type, choice(chosen) tree structure specified for the nested logit model type N trigger N k --------------------------------------------- no_trigger 38484 --- without 38484 12828 trigger 31626 --- accounting 17424 5808 +- rating 14202 4734 --------------------------------------------- total 70110 23370 Running the following simple nlogit command prompts the following error: . nlogit chosen leverage current_rating || type: maturity amount || trigger:, noconst case(loan_id) variable trigger has replicate levels for one or more cases; this is not allowed What does that mean? The dataset seems to be well defined - just as in the Stata help example. First I thought there might be some sort of identical observations. But including variables that differ either within "observations" of the same loan_id and/or differ between observations of different loan_ids did not help. Do you have any ideas or suggestions? Thanks Markus * * 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/