Statalist The Stata Listserver


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

st: RE: -heckman-


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: -heckman-
Date   Fri, 20 Oct 2006 13:08:18 +0200

Nishant:
Not only will the output tell you the number of observations, but also which
were censored and which aren't. The censored cases do provide information 
used in estimation. Especially in estimating the selection equation, how could 
you estimate the effect of variables on the probability of being selected if you 
ignored all non-selected cases, i.e. if your data consists only of selected 
cases. So the censored cases shouldn't be deleted/ignored.
HTH,
Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology 
Vrije Universiteit Amsterdam 
Boelelaan 1081 
1081 HV Amsterdam 
The Netherlands

visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z434 

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------

--- Nishant Dass wrote: 
heckman Change_WRTm4  myvar $controls i1-i48 y1-y13,
twostep select(dummym1m3 = $FIRMm4 roa_m4 roa_m5 roa_m6)

The number of observations in this regression appears to be
around 6,000 but my "dummym1m3" (i.e., the dep. var. of the
selection equation) has only half as many 1's. 

So, my question is - does Stata report the number of 0's
and 1's?  I was expecting to see only the number of
"selected" observations.

 

*
*   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