Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Panel and IV


From   "Schaffer, Mark E" <[email protected]>
To   <[email protected]>
Subject   st: RE: Panel and IV
Date   Wed, 30 Sep 2009 16:00:07 +0100

Philomena,

> -----Original Message-----
> From: [email protected] 
> [mailto:[email protected]] On Behalf Of 
> [email protected]
> Sent: Monday, September 28, 2009 10:12 AM
> To: [email protected]
> Subject: st: Panel and IV
> 
> Dear all, 
> 
> in my panel data i tested for heteroscedasticity and 
> autocorrelation using the -xttest3 and -xtserial commands. 
> Both is present. I have now some few questions:
> 
> 1. -xtivreg2 worked well with the robust option, but how can 
> i correct for autocorrelation?

If by "correct for" you mean obtain SEs that are robust to within-panel
autocorrelation, or you mean obtain coefficient estimates that are
efficient in the presence of arbitrary within-panel autocorrelation, you
can use -xtivreg2- with the cluster option, and with the cluster and
gmm2s options, respectively.

> 2. Is there a reset test for panel data (and instrumental variables)?

-ivreset- doesn't work with panel data (at least, not yet - it's
something I've thought about).  But if you are using first differences
or fixed effects, you can first transform your data "by hand".  For
example, if you are using an FE model, then after demeaning your data by
hand (say, using Ben Jann's -center- command) you can estimate the model
using -ivreg2- or -ivregress- and then call -ivreset-.

The only thing you need to be aware of is that small sample statistics
you get for the FE estimator done by hand will need a dof correction for
the fixed effects.  This is discussed in standard textbooks, so it
shouldn't be too hard to find.

Cheers,
Mark

> 3. Is there a way to test for autocorrelation and 
> heteroscedasticity when using -xtivreg (panel and IV)
> 
> 
> Any comment will be highly appreciated. Thank you very much, 
> 
> Philomela
> 
> 
> 
> -- 
> Jetzt kostenlos herunterladen: Internet Explorer 8 und 
> Mozilla Firefox 3.5 -
> sicherer, schneller und einfacher! 
> http://portal.gmx.net/de/go/atbrowser
> *
> *   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index