Statalist


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

Re: st: Changes to Maximize and Newton-Raphson in Stata 11?


From   jpitblado@stata.com (Jeff Pitblado, StataCorp LP)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Changes to Maximize and Newton-Raphson in Stata 11?
Date   Thu, 17 Dec 2009 12:35:14 -0600

Alfonso Miranda <a.miranda@ioe.ac.uk> posted that he was having some version
control issues with -ml-.

He sent me some files that represented the problem he was experiencing.

In Stata 11, -ml- was changed to be consistent with Mata's optimizers.  This
means that method 'd2' evaluators are now expected to return a Hessian matrix
instead of the negative Hessian matrix.

However -ml- has the -negh- option which indicates that the method 'd2'
evaluator is going to return a negative Hessian matrix.

Those of you who want to call -ml- under -version 11-, but want to continue
using your existing method 'd2' evaluator without change, will need to specify
the -negh- option.

We should have emphasized this in -[U] 1.3 What's new-, but didn't.

--Jeff
jpitblado@stata.com
*
*   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