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