Re: Problem with least-squares fitting.

[Tiago de Paula Peixoto <count0 at fissionauthority dot com>]

On Tue, 5 Nov 2002 13:42:27 -0600
Atakan Gurkan <ato@northwestern.edu> wrote:

> himm, i agree that n=1 should be solveable, but for a single parameter,
> the method might not be the most efficient. eg.
> y = a*x^8 -> x=(y/a)^(1/8), y  = a * sin(x) -> x = arcsin(y/a) ie.
> analytic approach. or if it is not possible: y = f(x) is equivalent to
> y-f(x)=0 which is solveable by root finding.

Yes sure, but the program that I'm writing fits a function provided by the
user. It should be able to fit any function provided, with any numbers of
parameters. The program reads the function, parses it and uses multifit to
fit them. It actually works with one parameter only (apparently), only it
displays this annoying message.

I could solve analytically like you suggest when the supplied function
only has one parameter, but if this is a bug in the GSL solver it should
be fixed, right? I don't care much about efficiency in this case. Having
the same code working for all cases is what I want.

-- 

    Tiago

Attachment: msg00107/pgp00000.pgp
Description: PGP signature