Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: RE: xtivreg2, robust error


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: xtivreg2, robust error
Date   Mon, 2 Jul 2012 15:56:03 +0100

Ari,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ari Dothan
> Sent: 02 July 2012 11:47
> To: statalist@hsphsun2.harvard.edu
> Subject: st: xtivreg2, robust error
> 
> Hi all,
> I use xtivreg2 regressions specifying "fe gmm2s bw(3)".

This means "robust to autocorrelation but NOT heteroskedasticity",
whereas

"fe gmm2s bw(3) robust"

means "robust to autocorrelation AND heteroskedasticity".

> In some cases I get more efficient results (smaller standard 
> error) when specifying "fe gmm2s bw(3) robust" than without 
> the "robust"
> requirement.
> How could this be explained?

You can't safely interpret the differences in this way.  The theoretical
results for efficiency of the 2-step GMM estimator are asymptotic, i.e.,
as the sample size goes to infinity.  What happens in the real world,
with finite samples, can be different.

HTH,
Mark

> Thanking you
> 
> --
> Ari Dothan
> *
> *   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

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   |   Site index