Mathematica 9 is now available
Student Support Forum
-----
Student Support Forum: 'NonlinearRegress merit function value' topicStudent Support Forum > General > "NonlinearRegress merit function value"

Next Comment >Help | Reply To Topic
Author Comment/Response
David Proctor
11/09/02 02:08am

I am trying to characterize the chi^2 merit surface for the non-linear fitting of a set of data to a sum of exponential decays:

a*e^(t/tau1)+b*e^(t/tau2)+c

I do this by providing a list of vaules for tau1 and fitting for everything else with NonlinearRegress, and plotting the final chi^2 value from the individual fits as a function of tau1. I've been calculating the chi^2 values with a separate statement (squaring and summing residuals) because I can't figure out how to extract the final chi^2 value from Mathematica's fit. Does anybody know how to do this? I'd like to avoid the separate statement because it slows the computation (the data set is ~12000 points long).

Thanks,
David

URL: ,

Subject (listing for 'NonlinearRegress merit function value')
Author Date Posted
NonlinearRegress merit function value David Proctor 11/09/02 02:08am
Re: NonlinearRegress merit function value Henry Lamb 11/12/02 02:39am
Re: NonlinearRegress merit function value David Proctor 11/14/02 1:06pm
Next Comment >Help | Reply To Topic