Statalist The Stata Listserver


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

Re: st: RE: -heckman-


From   Nishant Dass <nishant_dass@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: -heckman-
Date   Fri, 20 Oct 2006 05:05:13 -0700 (PDT)

Hi Maarten,

Thanks for your reply.  

I do understand that the selection equation has to include
the censored observations also in order to estimate the
hazard rate, but my question was about the 2nd stage
("outcome" equation) - that should only use the "selected"
observations, right?

So with that in mind, I thought the no. of obs. should only
show the no. of "selected" obs. used in the 2nd stage.  

I only wanted to make sure that the no. of obs. "displayed"
includes both, "selected" and "censored" obs, and that
there's nothing wrong with my understanding.  

Nishant

--- Maarten Buis <M.Buis@fsw.vu.nl> wrote:

> 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/
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
*
*   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