Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Second derivative for GMM


From   John Bates <lirac271@yahoo.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Second derivative for GMM
Date   Mon, 11 Jul 2011 11:33:23 -0700 (PDT)

Maybe I'm just confused, but I always thought the problem was to minimize the Objective Function E[xu] (where x are covariates and u is the residual).  It is only equal to zero in the just-identified case.  To minimize it, you would take the first derivative with respect to the parameters and set those equal to 0.  But then you would need the second derivative for the Hessian, right?



----- Original Message -----
From: Stas Kolenikov <skolenik@gmail.com>
To: statalist@hsphsun2.harvard.edu
Cc: 
Sent: Thursday, July 7, 2011 7:24 PM
Subject: Re: st: Second derivative for GMM

On Thu, Jul 7, 2011 at 10:43 AM, John Bates <lirac271@yahoo.com> wrote:
>   Stata's GMM command allows the user to specify analytic first derivatives. Is there any way to specify the second derivatives?  (Or will this functionality be added in future versions?)  I would think that would speed the command up.

I did not think the second derivatives play any role either in the
theory of GMM or in numerical procedures. You are minimizing the GMM
objective function; the Jacobian is a linear combination of the moment
conditions, and the Hessian is the linear combination of the first
order derivatives. Where do the second derivatives come into play?

-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   For searches and help try:
*  http://www.stata.com/help.cgi?searchhttp://www.stata.com/support/statalist/faqhttp://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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index