This is the mail archive of the gsl-discuss@sources.redhat.com mailing list for the GSL project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: constrain on nonlinear least-squares fitting

From: Brian Gough <brian dot gough at network-theory dot co dot uk>
To: "Fiaschi" <blazar at virgilio dot it>
Cc: <gsl-discuss at sources dot redhat dot com>
Date: Fri, 24 Jan 2003 20:32:57 +0000
Subject: Re: constrain on nonlinear least-squares fitting
References: <Pine.GSO.4.33.0301150926550.1569-100000@pollux.usc.edu><15909.53375.599774.657130@hppav.local><003501c2bdae$8364b690$0100a8c0@athlon>

Fiaschi writes:
 > How I can limit the range of one or more parameters on nonlinear
 > least-squares fitting algorithm?
 > 
 > I must fit a one-dimensional array with a simple generic sinusoidal function
 > I(t)=B+A*sin(omega*t+phi) but the parameters A and B must be always >0 and
 > 0<phi<pi

GSL doesn't have any routines for constrained minimisation.  The only
way is to use a penalty function with a large value outside the
desired region.

References:
- Bug in discrete random distribution
  - From: ahoward
- Re: Bug in discrete random distribution
  - From: Brian Gough
- constrain on nonlinear least-squares fitting
  - From: Fiaschi

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]