RE: Nonlinear fit
- To: mathgroup at smc.vnet.net
- Subject: [mg41116] RE: [mg41096] Nonlinear fit
- From: "Yahoo" <stefanofricano at yahoo.it>
- Date: Sat, 3 May 2003 03:30:25 -0400 (EDT)
- In-reply-to: <200305020759.DAA15079@smc.vnet.net>
- Reply-to: <fricano at dsfa.fisica.unipa.it>
- Sender: owner-wri-mathgroup at wolfram.com
The problem is, probably, in the definitions of D[f[x],p1] and respect to p2 etc... There are some e-mail on Mathgroup database about it, see them. *********************************************** Stefano Fricano Dip. di Scienze Fisiche ed Astronomiche Universit\140 degli Studi Via Archirafi, 36 Palermo 90123 Tel. +39 091 6234207 Fax +39 091 6168461 ********************************************* ->-----Original Message----- ->From: Gustavo Seabra [mailto:seabra at ksu.edu] To: mathgroup at smc.vnet.net ->Sent: Friday, May 02, 2003 10:00 AM ->To: mathgroup at smc.vnet.net ->Subject: [mg41116] [mg41096] Nonlinear fit -> -> ->Hi all, -> ->I am trying to make a nonlinear fit to a function (f[x]) that is defined ->differently for different intervals, something like (it's ->abbreviated here, ->to give just the idea): -> f[x_]:=g[x+a]/;x>k -> f[x_]:=g[x+b]/;x<=k -> ->I know the function is ok because, if I set the parameters, I can plot it ->with no problem: -> Plot[f[x]./{p1->p1_initial_value,p2->p2_initial_value,...},{x,x1,x2}] -> ->However, if I try to fit a data set to this function with NonlinearRegress ->or NonlinearFit: -> NonlinearFit[data,f[x],x,{p1,p2,...}] -> ->I get: ->"NonlinearFit::nonum : -> The model is not numerical at {p1->value,p2->value...,x->value}. Check ->that all model parameters are included in the parameter list." -> -> Now, I'm sure I included all parameters. As I said, if I try to plot ->f[x], even with the parameters from this answer, it works fine. Also, if i ->use a different f[x], defined the same for all x, all works fine. Any ->suggestion on what am I doing wrong? -> -> Thanks a lot, -> Gustavo Seabra. -> ->P.S. If anyone needs more detail, I can provide the notebook with ->everything ->in it. I just didn't want to post it here to avoid too long a post. -> -> -> ->
- References:
- Nonlinear fit
- From: "Gustavo Seabra" <seabra@ksu.edu>
- Nonlinear fit