Statalist The Stata Listserver


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

RE: ivhettest: heteroskedastic IV in panel estimation


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: ivhettest: heteroskedastic IV in panel estimation
Date   Mon, 24 Jul 2006 12:56:22 +0100

Vincenzo,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Vincenzo Lombardo
> Sent: Sunday, July 23, 2006 10:22 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: ivhettest: heteroskedastic IV in panel estimation
> 
> dear all,
> i am using this commands (ivreg2, gmm) for my panel 
> estimation; my issue is this one: i have to assess if using 
> IV estimation, that is I have basically to perform a test of 
> endogeneity. Now, in my panel there is presence of 
> heteroskedasticity and serial correlation (within cluster and 
> over time); then I tried two basic way to deal with this:
> 1) estimating the classical iv panel model re (xtivreg, re), 
> estimating the model not instrumented (xtreg, re) and then 
> performing the hausman-type test on the two models; results 
> not need to instrument!!!

But if there is heteroskedasticity and autocorrelation, then xtreg,re is
not efficient, so the Hausman test is invalid.  Or am I missing
something?

> 2)then, taking into account the fact the the above 
> estimations do not consider the presence of 
> heteroskedasticity and correlation (xtivreg and xtreg are not 
> allowed for this), i am using ivreg2, gmm (because it has 
> been showed that in this cases GMM is more efficient than IV) 
> with robust-cluster option.

Shouldn't you use xtivreg2 instead of ivreg2?  The former is a wrapper
for the latter but gets things like degrees-of-freedom adjustments
right.  Or you can use ivreg2 but with explicit fixed effects dummies.

> 3) QUESTION: can I use after step 2, above, the command 
> ivhettest to demostrate the presence of heteroskedasticity in 
> the second model, even if my dataset is panel?

Actually, this is a reason to use ivreg2 with explicit dummies.
ivhettest won't work after xtivreg2, but will work after ivreg2 with
dummies.

HTH.

--Mark

Prof. Mark E. Schaffer
Director
Centre for Economic Reform and Transformation
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://www.sml.hw.ac.uk/cert
 

> finally, do you think that the step 1 can give consistent 
> result of the endogeneity test, that i did?
> thanks a lot for your attenction,
> warm regards, vincenzo.
> 
> 
> --
> Vincenzo Lombardo
> University of Sussex, Dept. of Economics University of Napoli 
> Parthenope, DSE
> 12 Canfield Road
> BN2 4DN Brighton - UK
> tel. (UK):++44(0)7765991711
> tel. (ITALY):++393284164302
> *
> *   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/
> 
> 

*
*   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