Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | <econman200@earthlink.net> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | st: simulated annealing |
| Date | Thu, 3 Oct 2013 15:14:28 -0400 (GMT-04:00) |
I have a user defined simulated likelihood function that fails to converge under the standard optimization algorithms (e.g., nr, bfgs, etc.). I have also read that simulated annealing can be used to obtain convergences in instances where traditional methods break down.
Does anyone know if Stas Kolenikov’s Mata function -simann- has been integrated with any of Stata’s likelihood evaluators? Or, is there a publicly available/shareable likelihood function that relies on simulated annealing?
David
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/