Re: [R] using user-supplied derivatives in rgenoud

From: Uwe Ligges <ligges_at_statistik.uni-dortmund.de>
Date: Thu 02 Jun 2005 - 19:39:04 EST

Thoralf Mildenberger wrote:
> I have been using the rgenoud package for a nonlinear least-squares
> problem with lots of local minima, and it works very well but takes lots
> of time. According to the article refrenced in the documentation, the
> original GENOUD-software by the same authors seems to allow for
> user-supplied analytical derivatives instead of numerical approximations,
> which would probably save some time. Does anybody know whether this
> feature is also available in rgenoud? The documentation says nothing about
> this, but rgenoud seems to call optim() in the stats-package, which does
> allow for passing a function evaluating the derivatives.
>
> Thank you,
> Thoralf Mildenberger

Looking at the code suggests that you cannot supply derivates. Anyway, this question is very special and should perhaps be addressed to the maintainer of the package rather than R-help.

Uwe Ligges

> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Thu Jun 02 19:51:37 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:32:21 EST