Statalist The Stata Listserver


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

Re: st: two endogenous regressors


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: two endogenous regressors
Date   Fri, 24 Mar 2006 17:05:11 -0500

That F>10 "rule of thumb" from Staiger and Stock (1994) does not apply
to even the case of one endog RHS variable and one instrumental
variable, now that more research, e.g. Stock and Yogo (2002), has been
done on the topic.  (Note that -ivreg2- reports the required
Cragg-Donald statistic when option ffirst is specified--thanks again,
Mark Schaffer!)

Staiger, Douglas and Stock, James H., "Instrumental Variables
Regression with Weak Instruments" (January 1994). NBER Working Paper
No. T0151. http://www.nber.org/papers/T0151

Stock, James H. and Yogo, Motohiro, "Testing for Weak Instruments in
Linear IV Regression" (November 2002). NBER Working Paper No. T0284.
http://www.nber.org/papers/T0284

but see
http://ksghome.harvard.edu/~JStock/ams/websupp/index.htm

On 3/24/06, NEYMOTIN, FLORENCE <neymotin@calmail.berkeley.edu> wrote:
> Hello, does anyone know the correct test of the first stage when I have two endogenous regressors
> and two instruments? (so my equation is exactly identified).  I heard something about an S-Test,
> and I'm not exactly sure how my "rule of thumb" about an F-stat of at least 9 or 10 for the first
> stage changes when I have multiple first-stages to worry about.

*
*   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 1996–2020 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index