Statalist


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

st: heckman twostep selection model - not using all available observations


From   "georg wernicke" <georg.wernicke@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: heckman twostep selection model - not using all available observations
Date   Tue, 7 Aug 2007 17:26:13 +0100

Hello,

I have probably rather easy question but which is just driving me crazy.

I have a dataset with bilateral Investment flows. Unfortunately, the
dependent variable is quite often missing, negative or zero. The
respective literature suggests to use the Heckman selection procedure
(twostep) in order not to loose all the information where the
dependend variable is missing, zero or negative. I created an dummy
which is equal to one when there is positive investment flow and zero
otherwise (missing, zero, negative). I ran the heckman twostep
regression. However, stata is not using all possible observations.
Even worse, out of around 4100 observations, stata uses only 874 and
cencores 722 and only does not cencore 152. But I have 759 positive
observations for my dependent variable (where the dummy=1). Why is
stata not using 759 obs as uncencored?  why is stata not using all
observations? I have checked all available sources, but nobody seems
to have had the same problem.

Your help is most appreciated.

Sincerely,

Georg Wernicke
*
*   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