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   Alfonso Miranda <a.miranda@ioe.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Changes to Maximize and Newton-Raphson in Stata 11?
Date   Wed, 16 Dec 2009 19:35:19 +0000

Dear All,

I also have been experiencing "problems" with the NR method algorithm in
Stata 11. I have some ML routines that work fine in Stata 10.1. However, as
soon as I eliminate the Version 10.0  control line at the beginning of my
code ML cannot find concavity anymore and keeps iterating forever. So, I
agree, ML in Stata 11 seems to have gone mayor changes and things that work
in Stata 10 do not seem to work the same way in Stata 11.

Is there any explanation for this behaviour?

Best wishes,

Alfonso.


On 16/12/2009 19:24, "Brooks Taggert J" <brooks.tagg@uwlax.edu> wrote:

> Statalisters-
> 
> So I ran across a "problem". I have Stata 11.0 but I always use version
> control in my do files to avoid breaking the ability of the file to run
> through. Anyhow, I was estimating a stereotype logit -slogit-, and it
> worked fine using the syntax in interactive mode, but when running the
> do file it would fail to converge. It was maddening that it seemed to
> work when I cut and pasted the same command into the command dialog from
> the do file, but the do file would hang. Eventually I figured out this
> was the fault of the Version 10.0 at the beginning of the do file. So
> clearly there must have been some changes to the NR method or maximize
> more generally, but I can't seem to find reference to it on the Stata.
> Anyone have an idea on this? I just want to make sure the version 11.0
> result is the one that should be suspect.
> 
> Thanks
> 
> TJ
> 
> Taggert J. Brooks, PhD
> Associate Professor
> Department of Economics
> University of Wisconsin - La Crosse
> 
> 
> 
> 
> *
> *   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/

----------------------------------------------------------
Dr. Alfonso Miranda
Lecturer in Quantitative Social Science

Institute of Education, University of London.
20 Bedford Way, London WC1H 0AL, UK.
Tel. +44 (0) 020 7331 5105
E-mail: A.Miranda@ioe.ac.uk







*
*   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