st: selection with panel data - which model would work?

["Nora Reich" <nora11@gmx.de>]

Dear Stata users,

I am searching for a method on how to do the selection into employment with panel data, so that in the second step I can estimate the wages of the employees.

I found that some people use rather simple estimations. One of them estimates a Heckman selection and then sets the data to stset, similar to

probit working indepbvar
predict selection,xb
tsset id_ind wave
tsreg lnwage indepvar selection

Another one used the standard cross-sectional Mills ratio formular for panel data, similar to this:

xtprobit working indepvar   /* or xtlogit*/
predict p1, xb  /* from UCLA doc
generate phi = (1/sqrt(2*_pi))*exp(-(p1^2/2)) /*standardize it*/
generate capphi = norm(p1)
generate invmills = phi/capphi
xtreg lnwage indepvar invmills


My first question is, would these "easy" methods work? I know there has been a discussion on the second suggestion (http://www.stata.com/statalist/archive/2005-06/msg00426.html) but at that point of time, users had disagreed whether this estimation can be used.

I also found some other hints in this Stata user list, like using gllamm, or implementing models suggested by

Kyriazidou, E. (1997). Estimation of a Panel Data Sample Selection Model.
Econometrica. 65 (6), pp. 1335-1364.

Wooldridge, J. (1995). Selection Corrections for Panel Data Models under
Conditional Mean Independence Assumptions. Journal of Econometrics. 68, pp.
115-132.

However, it is difficult for me to apply theoretical models in Stata. Hence, my second question is whether, since the last posts, there are examples for the commands of selection models using gllamm, Kyriazidou's or Wooldridge model in Stata.

Best Regards
Nora


-- 
GRATIS! Movie-FLAT mit über 300 Videos. 
Jetzt freischalten unter http://portal.gmx.net/de/go/maxdome
*
*   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/