Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Part II: An error in the xtabond-description


From   Giovanni Bruno <giovanni.bruno@uni-bocconi.it>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Part II: An error in the xtabond-description
Date   Mon, 23 Aug 2004 18:40:58 +0200

Tewodaj

I agree. 

Concerning your first message, in -both- one-step and two-step formulae one 
expects the -inverse- of the estimated var-cov matrices of moment conditions as 
weighting matrices. So, the typos regard the definition of A1 at page 30 and 
that of A2 at page 31. The correct definitions are obtained by taking the 
inverse of the right hand side cross-product matrices.

As to the second message, one-step residuals enter the criterion function 
minimised by the two-step estimator, but the Sargan-test formula actually 
requires two-step residuals.

I hope this helps
Giovanni 


TEWODAJ MOGUES <tmogues@students.wisc.edu>:

> Hi, 
> 
> A quick follow up to my earlier message. There seems to be another error
> (please let me know if I'm wrong, anybody) in the Stata manual's xtabond
> description. This time it concerns the Sargan test on p. 32 bottom. The
> weight in the Sargan test statistic S2 that the manual indicates is A2, which
> is defined on p.31 and involves the residuals from the one-step estimation. 
> 
> However, looking both in the original paper of Arellano and Bond 1991, as
> well as in Bond 2002 ("Dynamic Panel Data Models: A Guide to Micro Data
> Methods and Practice" ... by the way, this is a wonderfully intuitive
> treatment of estimation methods for dynamic panel data models, I would highly
> recommend it!), the weight should really involve the residuals from the
> two-step estimation, not from the one-step estimation. 
> 
> Thanks,
> Tewodaj
> 
>    Date: Sat, 21 Aug 2004 11:18:47 -0400
>    From: TEWODAJ MOGUES <tmogues@students.wisc.edu>
> Subject: st: Error in xtabond-description in the manual
> 
> Dear Statalisters (and Stata programmers)
> I don't know if I have not-the-newest Stata manual books, but in the one I
> have there is an error in the way that the mathematical expression of the
> Arellano-Bond estimator (programmed by xtabond) is written down. In the xt
> book, on page 31, The expressions for delta2-hat and for Q2 should include
> A2^(-1), and not A2.
> Tewodaj
> 
> 
> ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
> Tewodaj Mogues
> Dept. of Agricultural and Applied Economics
> University of Wisconsin - Madison
> 427 Lorch St. #317, Taylor Hall
> Madison, WI 53706
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 


-- 
Giovanni Bruno
Istituto di Economia Politica, UniversitÓ Bocconi
Via U. Gobbi, 5, 20136 Milano
Italy
tel. + 02 5836 5411
fax. + 02 5836 5438
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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