A discussion paper version can be found here:
http://www.cgdev.org/content/publications/detail/11619
And more specifically on the subject of too many instruments:
http://www.cgdev.org/content/publications/detail/14256
Both are by David Roodman, the author of xtabond2
Eric de Souza
College of Europe
Brugge (Bruges), Belgium
http://www.coleurope.eu
-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Martin Weiss
Sent: 02 May 2009 17:11
To: statalist@hsphsun2.harvard.edu
Subject: st: Re: xtabond2 questions: too many instruments / singular
covariance matrix
<>
There was a comprehensive introduction to this command recently in the
SJ
http://www.stata-journal.com/article.html?article=st0159
which may be worth reading for you...
HTH
Martin
_______________________
----- Original Message -----
From: "susie karnes" <susiekarnes@gmail.com>
To: "statalist" <statalist@hsphsun2.harvard.edu>
Sent: Saturday, May 02, 2009 4:58 PM
Subject: st: xtabond2 questions: too many instruments / singular
covariance matrix
> Dear statalisters:
>
> I am getting the following warnings when I run a system GMM with
xtabond2:
>
> Warning: Number of instruments may be large relative to number of
> observations.
> Warning: Two-step estimated covariance matrix of moments is singular.
> Using a generalized inverse to calculate optimal weighting matrix for
> two-step estimation.
> Difference-in-Sargan/Hansen statistics may be negative.
>
> Is it a concern that the matrix is singular? Is this because of the
> relatively large number of instruments here?
> What can be done to resolve the issue? Limiting the number of
> instruments and collapsing?
>
> Also, with the use of system GMM, is there a test that should be run
> that evaluates the stationarity of the series?
>
> Many thanks for any advice that you can offer.
> Best regards,
> Susan
>
> *
> * 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/
>
*
* 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/
*
* 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/