Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Daniela Stockmann <ds011i@yahoo.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Computing stats for fixed effects model after xtreg, fe |
Date | Sun, 15 Apr 2012 13:16:57 -0700 (PDT) |
I have just read a very helpful paper by Tom Clark and Drew Linzer, entitled "Should I Use Fixed or Random Effects?" The authors recommend to first fit a fixed effects model and to then compute the correlation between the estimated unit effects and the within-unit means of the independent variable. Do you have any suggestions about how to compute this correlation in STATA after xtreg, fe? The unit effect captures the amount by which predictions of the dependent variable in a unit (panel) must be adjusted upward or downward, given knowledge only of the independent variables. -- danie * * 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/