MathGroup Archive 2009

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

Search the Archive

Re: FindFit tips

  • To: mathgroup at smc.vnet.net
  • Subject: [mg99015] Re: FindFit tips
  • From: dh <dh at metrohm.com>
  • Date: Fri, 24 Apr 2009 03:45:52 -0400 (EDT)
  • References: <gspgq7$o1c$1@smc.vnet.net>


Hi Sean,

nonlinear fitting is not too easy. The problem comes from the fact that 

there can be many good fits, the error surface can have many local 

minima. To get the one you want, you may have to introduce constrictions 

of the parameters. E.g. p1>0.

Daniel





"Sean G." wrote:

> Hi, I am using FindFit to fit a function I have derived to a data set I have. However, FindFit returns garbage; I can do a better job fitting the set by hand.  I'm not smarter than mathematica, so I must be using FindFit wrong.  Any ideas?

> 




  • Prev by Date: Re: Change Figure according to Session Time
  • Next by Date: Re: Plot x to y and y to x
  • Previous by thread: Re: FindFit tips
  • Next by thread: fitting surface with Mathematica 6.0.3 and range of fitted values