Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: deriving and handling a selection bias


From   "Georg Wernicke" <georg.wernicke@googlemail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: deriving and handling a selection bias
Date   Tue, 25 Mar 2008 17:40:23 +0100

Dear stata-listers,

I am unsure how to derive and use the inverse mills. I used the Heckman
twostep built into Stata and the option mills(mymills) to save the mills.
Then I included the saved mills in my original OLS regression to see whether
or not it becomes significant. Is that a correct method to see whether a
selection problem is present (e.g. when the mills is significant in the
OLS)? 

Moreover, I have to use the cluster option which I cannot in the twostep
Heckman and the ML Heckman breaks down unfortunately.

Is there another way to compute the mills ratio and is my arguing in regards
to the presents of a selection process valid?

Many Thanks in advance

Georg

*
*   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/



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