Dear all statalisters,
I need some information about xtabond, and ivreg2.
Consider a dynamic panel regression as
(1) Y(it) = a + B(1it)* L.Y(it) + B(2it)* X(it) + a(i) + u(it)
Anderson and Hsiao (1981) suggest to first difference this equation to drop individual effect a(i). Arellano and Bond (1991) solve endogeneity problem applying GMM estimator and instrumenting LD.Y(it) with T-3 lags of dependent variable and T-2 lags of exogenous variables. Xtabond use as instrument T-p-2 lags of dependent variable in level, T-2 lags of exogenous variables in first-difference and T-p-1 lags of predetermined variables in -difference, where p is the number of dependent lag setting in equation (1).
The problem is that when i try to replicate this command using ivreg2 I can't use all these instruments as an increase in their number (number of lags) reduce the number of observation disposable for the second stage of regression. How can i solve this problem?
Thank you in advance for your usefull advices,
Best Regards
Andrea Sisto
University of Eastern Piedmont
____________________________________________________________
Libero ADSL: navighi gratis a 1.2 Mega, senza canone e costi di attivazione.
Abbonati subito su http://www.libero.it
*
* 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/
