The Heckman (or Crag) model has been extended to the panel setting, but mainly for fixed effects. RE selection models are difficult to estimate as there is no way to bring the individual effects forward from the first (probit or tobit) step. There is a growing literature on Heckman panel models. A recent contribution which gives a good introduction is Dustmann, C., & M. E. Rochina-Barrachina. "Selection correction in panel data models: An application to the estimation of females' wage equations." Econometrics Journal 10, no. 2(2007): 263-293.
I do not think estimation of panel selection models have been incorporated in commercial software.
regards,
Even Bergseng
>-----Original Message-----
>From: [email protected] [mailto:owner-
>[email protected]] On Behalf Of Anita
>Sent: Friday, February 08, 2008 11:09 AM
>To: [email protected]
>Subject: st: truncreg with panel data
>
>Dear Statlist members
>
>I would like to estimate a model based on Cragg (1971) which contains in
>a first step a probit estimation and in a second step a regression part
>for all positive outcomes of the probit estimation.
>I learned that I habe to estimate the second step with truncreg to
>account for the truncation. Unfortunately I am working with a panel
>dataset and I nowhere found a command integrating fixed/ or random
>effects. Is there a possibility to do it, when not, why?
>
>Thanks a lot
>*
>* 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/
