Statalist


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

Re: st: How low can the percentage of uncensored cases be in heckprob?


From   Steven Samuels <sjhsamuels@earthlink.net>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: How low can the percentage of uncensored cases be in heckprob?
Date   Tue, 11 Nov 2008 17:28:49 -0500

**

I see your point, Maarten. I would guess (being out of my element here) that the higher the percentage of uncensored observations, the more resistant will results in the second equation be to misspecification in the first.

--Steve



On Nov 11, 2008, at 5:00 PM, Maarten buis wrote:

--- Steven Samuels <sjhsamuels@earthlink.net> wrote:

I am not familiar with -heckprob-, but I doubt if the *percent* of
uncensored observations matters much.

  -heckprob- fits two probit models.

There is a bit more going on here, at the second equation you only have
information on those who where selected in the first equation, so
having many people censored means that you loose a lot of information
for the second equation.

-- Maarten

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

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

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



*
*   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/

*
*   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   |   What's new   |   Site index