MathGroup Archive 2003

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

Search the Archive

nonlinearfit

  • To: mathgroup at smc.vnet.net
  • Subject: [mg43422] nonlinearfit
  • From: Ming Hsu <mhsu at hss.caltech.edu>
  • Date: Fri, 29 Aug 2003 07:16:49 -0400 (EDT)
  • Organization: California Institute of Technology, Pasadena
  • Sender: owner-wri-mathgroup at wolfram.com

Hi, I believe this question was posted before, but I couldn't find any
followups with the answer.  I have a messy nonlinear regression where I
want to constrain a subset of the parameter values to be positive.  

The general problem is that there is some software slowdown problems that
I want to take account of.  I am adding a Exp[a*(t-1)] term to the
model (which is nonlinear anyways).  So I want to make sure that a>=0.  

I understand that nonlinearfit is based on findminimum, which I know has
that capability, but it doesn't seem to be an option for nonlinearfit. 

Thanks,
Ming 


  • Prev by Date: Using error message output
  • Next by Date: Re: How does MathReader work?
  • Previous by thread: Re: Using error message output
  • Next by thread: Discrete convolution problem