Maureen. xtivreg should give the same results as when you do the model by
hand. are you including all your exogenous variables (including the
excluded instruments) in the first-stage regression and are you including
the fixed effects in your prediction of the wage?
to demonstrate try:
use auto
xtivreg price gear_ratio displacement ( mpg = weight length), fe
i(foreign)
xtreg mpg gear_ratio displacement weight length, fe i(foreign)
predict fmpg, xbu
xtreg price fmpg gear_ratio displacement, fe i(foreign)
The coefficient results from the last command are identical to the xtivreg
results.
Hope this helps,
Steve
> -----Original Message-----
> From: Maureen Paul [SMTP:[email protected]]
> Sent: Thursday, July 31, 2003 12:05 AM
> To: [email protected]
> Subject: st: Stata/Econometric question
>
> Hi all
>
> I am estimating a labour supply equation with wage as a LHS variable. When
> I use -xtivreg, fe- the logwage variable becomes insignificant with a
> very high standard error. However, if I perform the IV manually by first
> estimating a fixed effects wage equation and then using the log of the
> predicted wage in the fixed effects hours equation, the estimate is fine.
>
> Note that the instrument used passes the tests of validity.
>
> My questions are
> 1. Can anyone tell me why I am getting these different results?
>
> 2. How much should I be concerned about this given that the wage is not
> really the variable I am interested in?
>
> Many thanks in advance for any help.
>
>
>
> *
> * 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/
The information contained in this document is intended only for the
addressee and is not necessarily the views nor the official
communication of the Department of Labour. All final/official papers
which are sent from the Department will be sent by non-electronic
means, on appropriate letterhead, signed by authorised personnel.
*
* 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/
