Mathematica 9 is now available
Student Support Forum
Student Support Forum: 'fitting-' topicStudent Support Forum > General > Archives > "fitting-"

Help | Reply To Topic
Author Comment/Response
07/27/07 09:57am

Version:5.0 for Win

While fitting to a nonlinear function (FindFit with initial parameter set) I got the following message:
FindFit::lstol: The line search decreased the step size to within tolerance \
specified by AccuracyGoal and PrecisionGoal but was unable to find a \
sufficient decrease in the norm of the residual. You may need more than \
MachinePrecision digits of working precision to meet these tolerances.
How should I proceed?
The first part of the fit, for small x values where y are big is OK, it is just the second one (y<0.001) where the fit strongly diverges. Before setting the initial parameters values I did a curve simulation -so I know I had to be pretty close to a local minimum, however the problem is there are 5 parameters with the dependecies close to 1, so the equation is definitely overparametrized. Nevertheless I should be able to fall into the closest local minimum, and the fit looks much worse than the simulated curve.
Second question, how can I, except from setting initial values of the parameters, put constraints on them? (I tried to follow the syntax I found in the help - did not work).
Thanks in advance,

best regards,

URL: ,
Help | Reply To Topic