Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Jochen Späth <jochen.spaeth@iaw.edu> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: -xi: regress- fails |
Date | Wed, 3 Nov 2010 20:09:55 +0100 |
Dear statalisters, I happen to have a mysterious problem with -xi- when trying to estimate a regression model with two dummy variables and interactions of various continuos variables and a dummy variable, i.e. xi: regress y i.dummy1 i.dummy2 i.dummy3*x1 i.dummy3|x2 ... i.dummy3|x7 Stata tells me "variable __000003 not found" and breaks the execution of -xi: regress-. After -set trace on- I recognized this being a problem of the _ggroup subroutine but at this point I don't know what to do. Should I simply rely on -tab z, gen(dummy)- and create all interaction terms manually? (however, for convenience, I would not prefer this workaround since I have many interaction terms.) Maybe I should mention that I am running -xi: regress- several times before I get the error message above and everything works fine. Many thanks, Jochen * * 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/