Mathematica 9 is now available
Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
2005
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*December
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 2005

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

Search the Archive

Re: nonlinear fit question

  • To: mathgroup at smc.vnet.net
  • Subject: [mg60856] Re: nonlinear fit question
  • From: "Ray Koopman" <koopman at sfu.ca>
  • Date: Fri, 30 Sep 2005 03:57:41 -0400 (EDT)
  • References: <dhgepn$1rd$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Kevin Towles wrote:
> I am trying to do a nonlinear fit using a model I have developed.  My
> problem is that my model is discontinuous and I only want to allow
> the NonlinearFit function to search a range for the value of one of
> the parameters. The actual implementation would be something like:
>
> NonlinearFit[data,Model[xVar,yParam],xVar,yParam]
>
> but I want yParam to be confined to a range as: ymin <= yParam <= ymax
>
> I tried using NonlinearFit[data,Model[xVar,yParam],xVar,{yParam,
> {ymin,ymax}}] but the function is obviously searching outside this
> range.  I know there is a good fit in the range that I am specifying,
> so I don't know why it insists on searching outside the start
> values.  Is there a way to confine my yParam?
>
> Any help is greatly appreciated.
>
> Kevin Towles

If you don't expect the optimizing value of yParam
to be at either boundary then try fitting tParam,
with yParam = ymin + (ymax-ymin)/(1+Exp[-tParam]).


  • Prev by Date: Re: Sorting Strings: How and Why so f$&!ing hard?
  • Next by Date: Re: Sorting Strings: How and Why so f$&!ing hard?
  • Previous by thread: Re: nonlinear fit question
  • Next by thread: Re: nonlinear fit question