[Date Index] [Thread Index] [Author Index]

Re: NonlinearFit

• To: mathgroup at smc.vnet.net
• Subject: [mg19665] Re: NonlinearFit
• From: "William M. MacDonald" <wm2 at umail.umd.edu>
• Date: Wed, 8 Sep 1999 02:24:14 -0400
• Organization: University of Maryland
• References: <7r24q2$6qd@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

In article <7r24q2$6qd at smc.vnet.net> , Virgil Stokes 
<virgil.stokes at neuro.ki.se>  wrote:

> In reference to Statistics `NonlinearFit` (vers. 1.4, author: John M. Novak)
> for Mathematica Version 3.0.
>
> I would like to be able to replace the minimization criterion
> that is currently in this package (least-squares). I am trying to
> find a different type of minimization criterion -- one that
> performs better (hopefully) for my particular problem.
>
> Is it possible to use this package such that one can override the
> current least-squares criterion with a user defined criterion?
>
> -- Virgil
>
>
I thought that all statistical minimized the sum of the squared differences
between data and the modelling function, although different techniques for
doing this exist.  So I don't understand your question.  But have you tried
Statistics`NonlinearRegression` which uses Levenberg-Marquard and can take
error bars into consideration?  It is very sophisticated.

--
William M. MacDonald
Professor of Physics
University of Maryland

Internet: wm2 at umail.umd.edu