s-news
[Top] [All Lists]

Re: Resource for Simulated Annealing

To: Alberto Murta <amurta@ipimar.pt>
Subject: Re: Resource for Simulated Annealing
From: Prof Brian D Ripley <ripley@stats.ox.ac.uk>
Date: Wed, 27 Feb 2002 18:40:47 +0000 (GMT)
Cc: Olivier FLORES <flores@cirad.fr>, "s-news@lists.biostat.wustl.edu" <s-news@lists.biostat.wustl.edu>
In-reply-to: <3C7D2659.1A36CE2B@ipimar.pt>
On Wed, 27 Feb 2002, Alberto Murta wrote:

> Olivier FLORES wrote:
>
> I want to use the Simulated Annealing method for function optimization.
>  I am using S+ with windows, but I don't know any function or library
> that uses this method. The  "optim" funtion running with R does.
> Is there any equivalent for S+ ? Is there some code available somewhere
> ?
>
>
> Have you considered porting the simulated annealing part of 'optim' to
> work with S-Plus? After all it's open source.

optim is written in R's internal language, inside two .Internal statements.
That's open-source (if not much documented), but the S equivalent is as far
as I know undocumented.  There's .Call in S-PLUS 6, but that is very
sparsely documented.

I'd love to see your examples of doing this sort of thing: please make them
available to us.

[MASS4 will have a version of optim for S-PLUS 6, but that will not be
released for a few months, and the simulated annealing method has not
as yet been ported: it may never be.]

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595


<Prev in Thread] Current Thread [Next in Thread>