Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: timeout option


From   jpitblado@stata.com (Jeff Pitblado, StataCorp LP)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: timeout option
Date   Wed, 19 Apr 2006 20:20:24 -0500

Sascha O. Becker <sascha.becker@gmx.de> points out -logit- and -probit-
do not use -ml- to fit their likelihood models:

> In reply to my question about a "timeout option", Jeff Pitblado gave a 
> very helpful reply (thanks!!) ...
> 
>  > In Stata 9, -bootstrap- has a -reject()- option and all -ml- based
>  > commands save -e(converged)- to indicate when they have converged to a
>  > solution. Combining them with the -iterate()- option can cut down on
>  > some of the time consuming bootstrap replicates.
> 
> ... but:
> Logit and probit, being ML procedures, do allow for -maximize_options-, 
> i.e. allow for -iterate()-, but do not seem to save -e(converged)- ?!

True.  I assumed Sascha was having trouble with commands that use -ml- to fit
the likelihood model.  Both -logit- and -probit- are very fast (having their
likelihood optimized in the executable instead of using -ml-), but used with
-bootstrap- suffer in the replicates if there are very few successes or
failures (relative to the sample size).

Unfortunately I do not have a solution for commands like -logit- or -probit-.

--Jeff
jpitblado@stata.com
*
*   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