Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Heckman Heterogeneity (endogeneity) Bias control how to do guide for STATA needed, SPSS ref. included


From   Dominik Becker <dominik.becker@wiso.uni-koeln.de>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Heckman Heterogeneity (endogeneity) Bias control how to do guide for STATA needed, SPSS ref. included
Date   Wed, 31 Mar 2010 14:20:48 +0200

The Stata commands are -heckman- for continuous outcomes and -heckprob-
for dichotomous outcomes in your equation of primary interest. As a
start, the respective help files should help you.

Best,
Dominik



Am Mittwoch, den 31.03.2010, 12:13 +0200 schrieb VUX VUX:
> Dear all,
> 
> I have a question that might have been already discussed here, as the problem I try to solve is a common one. I want to control for the endogeneity more specifically heterogeneity (not sample selection) bias with a Heckman procedure. 
> 
> Since I use SPSS and never touched STATA so far, I found a good guide how to do that with SPSS (good theoretical explanation of the 2 types of selection bias problems):
> 
> http://home.planet.nl/~smit9354/selbias/Heckman-SPSS.doc
> 
> However the guide also states, that there are practical limitations – which I unfortunately experienced:
> 
> Quote: Although theoretically the procedure sounds rather well, applying it in practice is not so straightforward. An important condition for its use is that the selection equation contains at least one variable which is not related to the dependent variable in the substantial equation. If such a variable is not present (and sometimes even if such a variable is present), there may arise severe problems of multicollinearity and addition of the correction factor to the substantial equation may lead to estimation difficulties and unreliable coefficients.
> 
> --> When running the adjusted regression the Lambda and my strategic dummy variable (which was significant before the Heckman procedure) are both insignificant and have a very high VIF (>100).
> 
> I need to get this Heckman correction working (get rid of the multicollinearity) to write my paper…Now I noticed that STATA has some pre-defined functions to correct a regression model by applying the Heckman procedure. Maybe I will get different results with the STATA procedure compared to the manually calculated SPSS procedure.
> 
> As I have never used STATA, it would be great if someone here could write me how to practically get this done with STATA.
> 
> Many thanks already!
> 
> 
> ps. in my first paper the strategic variable I want to control is a dummy variable (like in all the examples). However, in my second it is summed score, thus a continuous variable -> how can I do the Heckman control for such a variable?

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index