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

Re: Fit in Excel Mathlink

• To: mathgroup at smc.vnet.net
• Subject: [mg5005] Re: Fit in Excel Mathlink
• From: vvs124 at rsphy1.anu.edu.au (Victoria Steblina)
• Date: Sat, 19 Oct 1996 02:25:50 -0400
• Organization: Australian National University
• Sender: owner-wri-mathgroup at wolfram.com

In article <53sm1g$3nj at dragonfly.wolfram.com>
Harald Berndt <haraldb at nature.berkeley.edu> writes:

> Mark Dowell wrote:
> > 
> > Dear All,
> > 
> > Can anyone hellp I'm trying to fit some data under Excel with the Mathlink
> > addon the fitting equation is
> > 
> > y=a(0)*Exp(-a(1)*(x-400))+a(2)
> > 
> > I just need value for best fit in a(0),a(1),a(2). for a given range in Excel
> > cells can anyone help with the syntax
> 
> Well, the Standard Package "NonlinearFit" is supposed to do it, but I
> had very little luck with problems similar to yours. Maybe things will
> improve in Version 3.0


Same here. I did a lot of NonlinearFit with mathematica for 3 parameter
models...
It is VERY frustating because you can't cover ALL the parameter's
space, and you never really know HOW to reach the global minimum of
ChiSquared.
The only thing I can tell you, and it won't help much, I'm afraid, is
that you MUST have very good initial guess for your parameters. One of
the best methods, as far as I know, is the use of genetic algorithm to
obtain initial values.

Best of luck
Vicki
________________________
vvs124 at rsphy1.anu.edu.au