Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Sample selection on subsamples

From   "sebastian.blesse" <>
Subject   st: Sample selection on subsamples
Date   Fri, 3 May 2013 01:52:50 -0700 (PDT)

Dear Statalistusers,

my problem is the following: I have a panel (from 1999 to 2005,
observations are 410) over municipal expenditure data and several controls
and i want to run a difference in difference estimation on the expenditure
impacts of a reform taken place between 2000 and 2003. However, i want to
run regressions on subsamples. Firstly, the affiliation to a certain area
as it is indicated by a binary dummy variable. Secondly, on population
tresholds (<=2000, >2000) with respect to a certain base year, the status
of the administrative form in a given year (again, indicated by a binary
dummy). Finally, i want to run a regression on a subsample of treated
municipalities compared to the original comparison group of non-treated. In
all of these cases i use pooled OLS and fixed effects regressions with an
If option to indicate the subsample in use. Are there sample selection
issues to be aware of and if so how can i deal with them? The Heckman
selection model considers unobservable selection variables to my knowledge.
Any help would be appreciated.



View this message in context:
Sent from the Statalist mailing list archive at
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index