Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Antwort: st: re: Sargan test


From   Johannes Geyer <JGeyer@diw.de>
To   statalist@hsphsun2.harvard.edu
Subject   Antwort: st: re: Sargan test
Date   Fri, 18 Feb 2011 15:11:02 +0100

> <>
> Sargan test of overidentifying restrictions
>         H0: overidentifying restrictions are valid
> 
>         chi2(114)    =  120.7602
>         Prob > chi2  =    0.3145
> 
> 
> Sargan test of overidentifying restrictions
>         H0: overidentifying restrictions are valid
> 
>         chi2(117)    =  116.0779
>         Prob > chi2  =    0.5067
> 
> so, what could this tell you, and could we rely on these results.
> 
> It quite clearly indicates that you cannot reject the null, which is
> a good thing. The p-value would have to be less than 0.1 or 0.05 to 
> reject, so you are a long way from having any indicated problems 
> with the instruments. However, heed David Roodman's advice in 
> xtabond2 from SSC: the Sargan test is not very powerful when there 
> are many instruments. Not knowing how many observations you have, I 
> don't know whether 100+ instruments is a lot or a little.
> 
> Kit
> 

You can find some more details on this test in the pdf manual shipped 
with Stata 11. E.g. heteroskedasticity is also a problem. Note that you 
have to check the Ar2 test as well.

Two good introductory texts that came to my mind are 

http://ideas.repec.org/p/cgd/wpaper/103.html

and 

http://www.cemmap.ac.uk/wps/cwp0209.pdf

which cover the "Sargan test" or its robust version, the "Hansen test" 

johannes
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index