Is there a way to constrain the search for parameter values on fitnlm? I keep running into issues when trying to solve for 3 to 9 variables and on a single iteration one of the parameters takes a value that causes an error and the whole thing stops. Then I have to do a bunch of exploratory analysis to come up with a better starting value.
Thanks, Joe

Antworten (1)

Star Strider
Star Strider am 12 Dez. 2014

0 Stimmen

It is not possible to constrain the parameters in the Statistics Toolbox functions. The Optimization Toolbox function lsqcurvefit will allow you to limit the to ranges.

Kategorien

Gefragt:

am 12 Dez. 2014

Beantwortet:

am 12 Dez. 2014

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by