Statalist


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

st: applying the heckman two step selection process


From   "georg wernicke" <georg.wernicke@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: applying the heckman two step selection process
Date   Wed, 8 Aug 2007 17:43:11 +0100

hello statalist,

i have a problem with using the two step heckman selection process.

i have country paiwise flow data for investment. this is my dependent
variable. unfortunately, the dependent variable is quite often
missing, negative or zero. i thought about using the heckman two step
to account for that. however, when i employ the heckman, it does not
use all observations. i have over 4000 observations, but the heckman
just uses around 900 of them. also, the process should not cencor
around 700 observations (for which i have positive flows and which i
would like to use in the regression) but uses only 152 uncencored
observations and cencores only around 700 observation while neglecting
all the rest.

i checked every possible source to find out why it does that, but i
could not resolve the problem.

i would very much appreciate if one has an idea what could be my
mistake and knows any way to circumvent that problem.

thank you very much

georg wernicke
university college london
*
*   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