Statalist


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

st: Evaluating Instrument Strenght in the Arrelano and Bond (1998) GMM System Estimator


From   Stephen Armah <armah.stephen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Evaluating Instrument Strenght in the Arrelano and Bond (1998) GMM System Estimator
Date   Sat, 6 Jun 2009 18:50:55 -0500

Dear Statalist,
I have two quick questions
1) Arrelano and Bond (1998) state that the GMM System estimator needs
a panel with a large number of countries but the time periods may be
few. I have 31 countries and 6 years. Does the  Arrelano and Bond
(1998) GMM System Estimator with forward orthogonal differencing apply
to my case? I suspect endogeneity caused respectively by unobserved
and time-invariant country-specific effects and simultaneity.

2) How does one evaluate the strength of the instruments used in the
Arrelano and Bond (1998) GMM System Estimator using xtabond2.  With
ivreg2, its easy to get Cragg-Donald Wald Statistics to compare to
Stock and Yogo (2005) critical values. I have no idea how to  evaluate
the strength of the instruments using xtabond2. Note I am aware that
the Sargan, Hansen and Difference Sargan tell us about exogeneity.  I
am interested in evaluating  instrument strength not exogeneity.

Your help is appreciated.

Thanks and
be well.
*
*
*   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