Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: re: ivreg2 running OLS not 2SLS


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: re: ivreg2 running OLS not 2SLS
Date   Thu, 20 Sep 2007 11:23:50 -0400

Nirina--
If all your excluded instruments are dropped in the "first-stage"
regression, the IV estimate is the OLS estimate. If Z is collinear
with X, you do not have excluded instruments!

use http://fmwww.bc.edu/ec-p/data/wooldridge/mroz.dta
qui tab educ, gen(ded)
ivreg2 lwage exper expersq (educ=ded*)
reg lwage exper expersq educ

On 9/20/07, Kit Baum <baum@bc.edu> wrote:
> Nirina said
>
> I run ivreg2 and it is telling me that my endogenous variable is
> collinear with the instruments and then it is running OLS but still
> reporting Sargan test of overid.
>
> I find it wierd since this is what I would have expected it to do:
>
> drop the collinear variables from the list of  instruments and then
> still run 2SLS. Could anyone please help on what could be the trick?
>
> Per the Statalist FAQ, Nirina should show us her commands and the
> resulting output. Otherwise it is hard to conjecture what might cause
> the reported behavior.
>
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index