Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | rdae20@bath.ac.uk |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Interacting the treatment variable in treatreg |
Date | Sun, 18 Apr 2010 11:49:59 +0100 |
Hi,I?m trying to interact the treatment term, in a treatment effects regression, with another variable. I?m using stata 10.
My current specification can be simplified to the equation below, where y1 (binary) is the treatment, being interacted with x3 (continuous).
xt: treatreg x1 x2 i.y1*x3, treat (y1 = y2 y3)The model seems to work, however y1 is dropped from the estimation. A term _iy1_1 does feature in the output, which I think represents (/replaces) y1, however I am not sure whether that term is the standard y1 variable (from the data) or the y1 predicted by the treatment equation.
If anybody could help, would be much appreciated? Thanks, Richard * * 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/