Statalist


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

RE: st: RE: cragg-donal test for weak instruments with heteroskedasticity


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: cragg-donal test for weak instruments with heteroskedasticity
Date   Fri, 25 Jul 2008 17:31:42 +0100

Sara,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> sara borelli
> Sent: Friday, July 25, 2008 4:58 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: R: st: RE: cragg-donal test for weak instruments 
> with heteroskedasticity
> 
> Dear Professor Shaffer,
> 
> I am sorry to ask you further,and such a long question, but I 
> want to be sure that I understand correctly. Thus, from my 
> understanding, I have two options:
> 
> Either I run ivreg2 and get the The K-P statitsics, and then 
> compare it to the tabulated values of Staiger and 
> Stock(2002), Or I explicitly estimate the reduced forms for 
> the two endogenous variables, compute the F-tests for the 
> excluded instruments in each reduced form and compare those 
> to 10. Is this correct?

Not quite.  Either you compare the K-P statistic (which is heteroskedasticity robust) to the Stock-Yogo (not Staiger-Stock) tabulated values for the C-D statistic, as you suggest, or you compare K-P to the the Staiger-Stock value of 10.

Separate first-stage regressions do not provide an adequate test of identification - see Baum et al (2003) for an explanation (full reference details in -ivreg2- help).

A third possibility is to test for heteroskedasticity (see -ivhettest-) and hope you don't have it.  If you don't, then you can use the C-D stat with S-Y critical values.

> Since both options are not going to give me the "perfect" 
> answer, I wanted to do it in both ways, and see what happens, 
> but after installing ivreg2 and ranktest, I am not able to 
> get the first stage K-P statistics or the cragg donald 
> statistics. After running
> ivreg2 y x (y1 y2=z1 z2), ffirst robust
> Stata reports only the first stage F (which is not even 
> robust) and the shea statistics, nothing more.
> Is it possible that this happens because I do not have the 
> appropriate stata version to obtain K-P or Cragg Donald? I am 
> currently using Stata 8. I have also access to Stata 10, but 
> there ivreg2 does not work, and it seems that K-P is a 
> special feature of ivreg2 and not of ivregress. Is this what 
> is going on?

Not quite.  -ivreg2- certainly runs under Stata 10, but you need to make sure you install the latest version, and also -ranktest- (which is the command that calculates the K-P statistic).  You are right that the K-P statistic is not supported by official Stata -ivregress-.

-ranktest- requires Stata 9 or better, which is why your Stata 8 version won't report the K-P statistic.

Cheers,
Mark

> Thank you very much for your help
> Sara
> 
> --- Ven 25/7/08, Schaffer, Mark E <M.E.Schaffer@hw.ac.uk> ha scritto:
> 
> > Da: Schaffer, Mark E <M.E.Schaffer@hw.ac.uk>
> > Oggetto: st: RE: cragg-donal test for weak instruments with 
> > heteroskedasticity
> > A: statalist@hsphsun2.harvard.edu
> > Data: Venerd́ 25 luglio 2008, 15:28
> > Sara,
> > 
> > > -----Original Message-----
> > > From: owner-statalist@hsphsun2.harvard.edu
> > > [mailto:owner-statalist@hsphsun2.harvard.edu] On
> > Behalf Of
> > > sara borelli
> > > Sent: Friday, July 25, 2008 1:57 PM
> > > To: statalist@hsphsun2.harvard.edu
> > > Subject: st: cragg-donal test for weak instruments
> > with
> > > heteroskedasticity
> > > 
> > > Dear All,
> > > 
> > > I a estimating the following model:
> > > 
> > > Y= aX + bY1 + cY2 + u
> > > I am using standard errors robust and clustered at the
> > 
> > > neighborhood level;
> > > 
> > > X = vector of eogenous regressors
> > > Y1, Y2  are endogenous
> > > Z1 Z2 excluded instruments
> > > 
> > > I want to test weather my IVs are non-weak. I know
> > that in
> > > the case of a single endogenous variable I should use
> > the
> > > "rule of thumb" that the first stage
> > F-statistic >10 (Steiger
> > > and Stock 1997).
> > > But Stock and Yogo (2002) say to use the Cragg -Donald
> > 
> > > Statistics in presence of multiple endogenous
> > regressors. I
> > > know this statistics is not valid under
> > heteroskedasticity. I
> > > would like to know if there is a way to compute a
> > robust
> > > analog of the Cragg-Donal statistics in Stata
> > 
> > The Cragg-Donald statistic is a test statistic for the rank of a 
> > matrix.
> > Anderson (1951) derived an earlier and similar test based 
> on canonical 
> > correlations.
> > 
> > -ivreg2-, downloadable from ssc-ideas in the usual way, reports a 
> > generalization of these to the case of heteroskedastic and/or 
> > autocorrelated errors, due to Kleibergen and Paap.  The K-P stat 
> > reduces to C-D and Anderson tests in the iid case.
> > 
> > The K-P stat is implemented in Stata by -ranktest-, so 
> you'll need to 
> > install this as well.
> > 
> > > and in case if
> > > it would still be valid to compare it to the tabulated
> > values
> > > of Stock and Yogo(2002). Alternatively, would be wrong
> > to
> > > simply refer to the old "rule of tumb" that
> > the F-statistics
> > > should be at least 10, even in presence of multiple endogenous 
> > > variables?
> > 
> > As far as I know, critical values have not been tabulated 
> for non-iid 
> > case (and it would be hard to, since the form of 
> > heteroskedasticity/autocorrelation will make a difference 
> to how they 
> > perform).
> > 
> > Either of your suggestions seems sensible to me.
> > 
> > Cheers,
> > Mark (coauthor of -ivreg2- and -ranktest-)
> > 
> > Prof. Mark E. Schaffer
> > Department of Economics
> > School of Management & Languages
> > Heriot-Watt University
> > Edinburgh EH14 4AS  UK
> > 44-131-451-3494 direct
> > 44-131-451-3296 fax
> > http://ideas.repec.org/e/psc51.html
> > 
> > 
> > > 
> > > Thank you for any help
> > > Sara Borelli
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > >       Posta, news, sport, oroscopo: tutto in una sola
> > pagina. 
> > > Crea l&#39;home page che piace a te!
> > > www.yahoo.it/latuapagina
> > > 
> > > *
> > > *   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 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/
> 
> 
>       Posta, news, sport, oroscopo: tutto in una sola pagina. 
> Crea l&#39;home page che piace a te!
> www.yahoo.it/latuapagina
> 
> *
> *   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 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/



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