st: sub samples

[Kit Baum <baum@bc.edu>]

Jean writes "but isn't there a constraint there that the fixed  
effects have the same coefficients in both samples?"

Yes, there is, but strictly speaking they are not covariates, are they?

If you want to allow for these to vary as well,

webuse grunfeld, clear
gen byte postwar_years = year >1945
g company2 = cond(postwar_years,company,company+10)
foreach var of varlist mvalue kstock  {
  gen post_`var' = postwar_years*`var'
}
xtreg invest mvalue kstock post* ,fe clu(company)

With that formulation you cannot test for an intercept shift, but you  
can test for slope shifts.


Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/