Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
2007
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 2007

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

Search the Archive

Re: Fitting data to a curve

  • To: mathgroup at smc.vnet.net
  • Subject: [mg73314] Re: Fitting data to a curve
  • From: Bill Rowe <readnewsciv at sbcglobal.net>
  • Date: Sun, 11 Feb 2007 01:17:11 -0500 (EST)

On 2/9/07 at 2:17 AM, benjamin.friedrich at gmail.com (ben) wrote:

>fit = FindFit[data,
>a x^2 + b y^2 + c x y + d x + e y + f, {a, b, c, d, e, f}, {x, y}]

>Please note

>Fit[] = linear fit = easy&stable = no initial guess required
>FindFit[] = nonlinear fit = subtle issue = initial guess helps a lot

=46indFit does both linear and nonlinear fits. For the linear case
there is no need to provide an initial guess when using FindFit.
=46indFit will always return the same result as Fit for the linear
case. Hence, there is no real reason to use Fit any longer.

>Always check your fitting results graphically.

This is particularly good advice for nonlinear problems when it
is possible to do so.
--
To reply via email subtract one hundred and four


  • Prev by Date: free boundary condition
  • Next by Date: Re: a function containg MeijerG (limit behavior)
  • Previous by thread: Re: Fitting data to a curve
  • Next by thread: DSolve y'[t]==y[t-a]