The Mathematica Journal


About the Journal
Current Issue
Editorial Policy
Back Issues
Contact Information


A Gauss-Newton Method for Nonlinear Regression

Volume 1, Issue 2
Fall 1990
M. N. Ediger, U.S. Food and Drug Administration

We implement a Gauss-Newton method for fitting a specified nonlinear model to observed values by using iterative linear-regression steps to reduce the sum-of-squares error between the model and observations. A brief example of fitting observed laser energy measurements to a Gaussian model is provided.

About Mathematica 
© Wolfram Media, Inc. All rights reserved.