Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Significance of adjusted Heckman estimates


From   Barbara Engels <engels.ba@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Significance of adjusted Heckman estimates
Date   Sun, 12 May 2013 12:41:08 +0200

Dear Sina, 

as far as I know, the second stage output of the Heckman command yields the estimated coefficients for the whole population, taking account of the censoring in the data, just as you mentioned.

However, I am also interested in the estimated coefficients and hence the output variable conditional on selection. That is why I use the inverse Mills ratio to calculate the adjusted coefficients. However, I don't know how to check if these coefficients are significant. 

Any ideas on this?

Thanks in advance!

Cheers,
Barbara




On 10.05.2013, at 17:11, Ghotbi, Sina wrote:

> Dear Barbara,
> 
> I have worked with Heckman models, but I am not sure if I fully understand your concern.
> 
> As far as I know, when you run a Heckman model in the idea world, the coefficient reported in the second stage (not selection) are already 'adjusted' for selection bias, and the standard errors are already reported as well.
> 
> Let me know if I am missing the point of your question.
> cheers,
> Sina
> 
> On 2013-05-10, at 2:51 AM, Barbara Engels wrote:
> 
>> Dear Statalisters,
>> 
>> I am estimating a two-step Heckman selection model (using Stata 10). The adjusted coefficients are calculated according to 
>> 
>> heckman  dep  indep, twostep select(select=indep exlcusionrestriction) mills(mills)
>> 
>> predict select_xb, psel
>> 
>> gen delta = mills*(mills + select_xb)
>> 
>> gen b_indep1 = [dep]_b[indep1] - ([blfp]_b[indep1]*e(rho)*e(sigma)*delta)
>> 
>> ci b_indep1
>> 
>> However, I don't know if this adjusted coefficient is statistically significant . indep1 appears both in the regression and in the seleciton equation, being significant in the regression and insignificant in the selection equation. 
>> 
>> 
>> How can I tell if the adjusted coefficient on indep1 is statistically significant?
>> 
>> 
>> Any help is greatly appreciated!
>> 
>> Cheers, 
>> 
>> Barbara
>> 
>> 
>> 
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index