Statalist The Stata Listserver


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

Re: st: More than 3 endogenous regressors


From   "Austin Nichols" <[email protected]>
To   [email protected]
Subject   Re: st: More than 3 endogenous regressors
Date   Mon, 23 Apr 2007 11:15:55 -0400

Dipa--
Mark and Kit have already pointed out why there are no Stock-Yogo
"critical values" (though these are not critical values in the
traditional sense, which will be clear after a quick read of the
Stock-Yogo paper) in your -ivreg2- output.

This link is a 2004 revision of the 2001 paper that first promulgated
the tables of cirtical values:
http://ksghome.harvard.edu/~JStock/ams/websupp/rfa_7.pdf
which says e.g. (p.24): "In contrast, the boundary of the bias region
for Fuller-k [LIML] tends to zero as the number of instruments
increase, which agrees with the consistency of the Fuller-k estimator
under many weak instrument asymptotics." and "Comparing these two
plots shows that tests based on LIML are far more robust to weak
instruments than tests based on TSLS."

Please note that -ivreg2- can do LIML too.

Read
http://ksghome.harvard.edu/~jstock/pdf/rfb_6.pdf
for more details on the various estimators.

On 4/23/07, Sarkar, Dipa <[email protected]> wrote:
Hi,

Can someone plz help?
I am using ivreg2: I have much more than 3 endogenous regressors and similarly high number of excluded instruments. How to test for weak instruments then using ivreg2?? The Stock-Yogo values are not reported for above 3 regressors (100 instrumemts).

Thanks.

*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index