st: RE: Hansen's overidentification test interpretation in xtivreg2

[DE SOUZA Eric <eric.de_souza@coleurope.eu>]

The statistic is 3.569 and its p-value is of 0.3119 !
Therefore, the null is not rejected
The null is that J =0, that the overidentification restrictions are valid.


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 Ari Dothan
Sent: 23 February 2011 08:52
To: statalist
Subject: st: Hansen's overidentification test interpretation in xtivreg2

Hi all, Professor Baum,

I am trying to understand whether the J-stat should be large or small.
There is some information which appears to be contradictory:

1. From the help screen: " The joint null hypothesis is that the instruments are valid instruments, i.e., uncorrelated with the error term, and that the excluded instruments are correctly excluded from the estimated equation". Therefore, a SMALL  J-stat will indicate that instruments are valid.

However, in the 2007 guidebook by Baum, Scahffer & Stillman, a Hansen statistic of 0.3119 is presented on page 11. The explanatory text is:
"the statistic is now far from the rejection of its null, giving us the confidence that our instrument set is appropriate"

I would much appreciate an explanation of why are the instruments "appropriate" when they are not valid.

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/

*
*   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/