Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Ye, Jingjing" <jingjingy@mail.smu.edu> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st: RE: heckman correction with endogenous variables included |
Date | Tue, 8 Mar 2011 01:32:56 +0000 |
Hi, Sharon What I did is use 2sls in the second step (instead of the regular ols in Heckman's model) along with the IMR obtained from the first step (the probit model with employed dummy as the outcome). However, if you want exclusive restriction for the purpose of identification, I think you need an "extra" instruments in the probit model. BTW, it is not restricted to use 2sls in the second step, I think. You can use the estimation method you want to deal with endogeneity. And there is non(semi)-parametric way to cope the two problems as well, for instance, full-information maximum likelihood estimation procedure(the Discrete Factor Method). But I don't know much about it. Hope it helps, **************************************************** Jingjing Ye Department of Economics Southern Methodist University Dallas, TX 75275-0496 **************************************************** ________________________________________ From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of xueliansharon [xuelianstata@gmail.com] Sent: Monday, March 07, 2011 7:08 PM To: statalist@hsphsun2.harvard.edu Subject: st: heckman correction with endogenous variables included Dear all, I want to estimate a wage equation (wage=x'b+u), and I want to use Heckman correction procedure to correct for sample selection bias since only working persons' wages are observed. However, the x's in the wage equation include two endogenous variables x1 and x2, and I want to employ instruments I1 and I2 (here, I1 is the IV for x1, I2 is the IV for x2) to address the endogeneity problem. But I don't know how to employ instrumental technique and Heckman correction simultaneously. Could anybody tell me how to solve this problem? Thanks in advance. Best, Sharon -- View this message in context: http://statalist.1588530.n2.nabble.com/heckman-correction-with-endogenous-variables-included-tp6099761p6099761.html Sent from the Statalist mailing list archive at Nabble.com. * * 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/