<>
I am sticking my neck out on this one, but could you -bootstrap- the entire
estimation procedure to arrive at an SE in the absence of an analytical
solution?
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Anselm Mattes
Gesendet: Mittwoch, 16. September 2009 13:18
An: [email protected]
Betreff: st: generated regressor and GMM
Dear Statalist,
I am trying to estimate a GMM (Blundell-Bond) model where one of my
regressors has been estimated by a probit model in a first step. Wooldridge
(2002), chap. 6 points out that the estimates are consistent, but standard
errors need to be corrected. In appendix 6A (p. 139) he derives the
corresponding formulas. Does anybody know, whether this (or another
suitable) solution is implemented in Stata?
I am aware of that this question has already been posted, but that was some
years ago and there was no solution given at that time.
I appreciate any help!
Best regards,
Anselm
_________________________________________________________________________
Anselm Mattes
Diplom-Ökonom
Email: [email protected]
Tel: +49 7071 - 9896 - 16
------
Institut für Angewandte Wirtschaftsforschung, Tübingen
Ob dem Himmelreich 1
72074 Tübingen
Tel: 07071 - 9896 - 0
Fax: 07071 - 9896 - 99
http://www.iaw.edu
_________________________________________________________________________
*
* 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/