RE: st: Heckman (sample-selecton) models and tests of model "adequacy"

[Christer.Thrane@hil.no]

As I understand it, a significant rho means that the two equations are
correlated, and this justifies the Heckman procedure. If Rho is not
significant, the (simpler) two-part model will suffice.

Christer


                                                                                                                                                  
                      "Subhash Pokhrel"                                                                                                           
                      <Subhash.Pokhrel@urz.uni-hei        To:       <statalist@hsphsun2.harvard.edu>                                              
                      delberg.de>                         cc:                                                                                     
                      Sent by:                            Subject:  RE: st: Heckman (sample-selecton) models and tests of model "adequacy"        
                      owner-statalist@hsphsun2.har                                                                                                
                      vard.edu                                                                                                                    
                                                                                                                                                  
                                                                                                                                                  
                      05.04.2004 19:17                                                                                                            
                      Please respond to statalist                                                                                                 
                                                                                                                                                  
                                                                                                                                                  




Dear Chris,

I have similar problem. I wonder why the reviewer was not happy with the
significance of rho (i'm sure you did report it). Any comment on that?
Subhash.

-----Original Message-----

Date: Fri, 2 Apr 2004 12:51:51 +0200
From: Christer.Thrane@hil.no
Subject: st: Heckman (sample-selecton) models and tests of model "adequacy"

Hi,

A reviewer has asked me to perform some "model adequacy" tests of my
Heckman
selecton model, i.e. a probit and a OLS regression conditional on
"answering
yes" in the probit.

I interpret this as he or she wants som measures of "model fit" (other than
the standard output) and possibly some measures that have to do with
diagnostic tests and/or specification tests?

The -findit- command did not help me here, so therefore: Any help to carry
out some of these tasks is appreciated.

Christer

Christer Thrane, dr.polit.
Associate Professor, Sociology
Department of Social Science
Lillehammer University College
2626 Lillehammer, Norway
+ 47 61 28 81 70 (fax)
+ 47 61 28 82 47 (phone, work)
+ 47 61 25 68 44 (phone, home)
+ 47 92 29 54 39 (mobile)
E-mail, work: christer.thrane@hil.no
E-mail, home: ch-thran@online.no






*
*   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/




*
*   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/