MathGroup Archive 2003

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

Search the Archive

Strange behavior with Nonlinear regress

  • To: mathgroup at smc.vnet.net
  • Subject: [mg40876] Strange behavior with Nonlinear regress
  • From: "Nathan Moore" <nmoore at physics.umn.edu>
  • Date: Tue, 22 Apr 2003 06:50:20 -0400 (EDT)
  • Organization: University of Minnesota, Physics
  • References: <b70npo$9gd$1@smc.vnet.net> <200304100736.DAA24158@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Having trouble with nonlinear regress function tonight.  I have a data set
that I'd like to fit an exapansion to.  When I use nonlinear regress to fit
the parameters they all diverge to non-physical quantities.  The help menu
seems to infer that it is possible to specif bounds for the fitting
parameters, but when I do so the program seems to find a fit with solution
outside the interval I've specified..  An example of this,

have data set of {x,y} points.
    correctedtr;
then try to fit the data,
    NonlinearRegress[correctedtr,   n^(2 nu)/a, n, {{a, 10, 0, 20}, {nu,
0.4, 0.3, 0.5}}]
assuming here that a should be found on the interval [0,20], nu on the
interval [0.3,0.5]
output in real life is,
    BestFitParameters{a->29.2876,nu->0.583367},

what's going on??????  what are the parametrs outside the bounds I specify?

I'd love to know

NT Moore



  • Prev by Date: small program in Mathematica
  • Next by Date: Re: defining a transform from two lists
  • Previous by thread: Re: sum with lists
  • Next by thread: Re: sum with lists