st: xtregar,re with endogenous variable & instruments

["Daniel Schalling" <daniel.schalling@uni-weimar.de>]

Hi Statalist,

I want to transform a xtivreg,re model into a xtregar,re model, due to fact
that the xtivreg,re model suffers from autocorrelation (AR1).
xtivreg y x1 x2 (IV = z), re

My thoughts were to do a xtreg,re model with the endogenous variable as the
depend variable and the instruments and independent variables from xtivreg
as independent variable. Afterwards I want to predict the fitted values of
the endogenous variable.
xtreg IV z x1 x2, re
Predict IV_xb, xb

Then I want to include the fittedvalue of the instrument into the xtregar,re
model as an independent variable and run the regression.
xtregar y x1 x2 IV_xb, re lbi

This proceed  seems quite logic to me, but I'm not that familiar with an
autocorrelated panel analysis with instruments.

Could you please say me, whether my thoughts are right and I can deal in
this way with my problem, or give me any other suggestion?
Thank you very much, Daniel.

*
*   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/