st: RE: Kleinbargen Paap

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Karan,

It's a tricky question.  The Stock-Yogo critical values can be used with
the Cragg-Donald statistic, but the theory assumes iid etc.

The Kleibergen-Paap statistic generalizes the C-D stat to cover things
like heteroskedasticity, but it has only an asymptotic justification.
That is, you can use it for testing for underidentification - like the
original use of the C-D stat, Anderson's test, etc.  But as far as I
know, nobody has worked out the theory, critical values, etc., for using
it to test for weak identification.

The only thing I can suggest is to use the Stock-Yogo critical values
anyway and just be cautious about interpreting the results.

BTW, you refer to the first-stage F-stat.  In the special case of a
single endogenous regressor, the Kleibergen-Paap statistic is
numerically equivalent to a robust first-stage F-stat.

HTH,
Mark

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Karan Nagpal
> Sent: 27 March 2012 14:34
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Kleinbargen Paap
> 
> Suppose I have one endogenous variable and I am using two instruments.
> I suspect there's heteroskedasticity, so the Cragg-Donald 
> test doesn't work, right?
> 
> Hence, I rely on the Kleinbargen-Paap test statistic, 
> obtained using the ivreg2 command. Am I right in assuming 
> that the null of the KP test is "instruments are weak?"
> 
> The statistic for the (Kleibergen-Paap rk Wald F statistic) is 29.356.
> Do I compare this with the Stock-Yogo thresholds or with the 
> first-stage F-statistic, to test the null?
> 
> Thanks,
> Karan
> *
> *   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/
> 


-- 
Heriot-Watt University is the Sunday Times
Scottish University of the Year 2011-2012

We invite research leaders and ambitious early career
researchers to join us in leading and driving research
in key inter-disciplinary themes. Please see 

http://www.hw.ac.uk/researchleaders

for further information and how to apply.

Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   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/