Create a Forum Topic by Logging In or Creating an Account

Curve fitting with multiple dependent and independent variables

Thu, 04/18/2019 - 08:32 pm

I was hoping to perform a custom nonlinear curve fitting of the form (y₁, y₂) = f(x₁, x₂; a₁,...,a_n), where x₁, x₂ are independent variables, y₁, y₂ are dependent variables, and a₁,...,a_n are fitting parameters. Any suggestions on how to leverage the existing Igor functions to achieve this purpose? It seems that

FuncFitMD

only accepts one dependent variable. Thanks!

johnweeks

Well, if you had only one independent variable, I would suggest our Global Fit package (see File->Example Experiments->Curve Fitting->Global Fit Demo). With multiple independent variables, you will have to roll your own.

The technique is to come up with a packing of your multiple independent variables into a single Y wave. Then you must write your fitting function to return the results organized appropriately. It is very difficult to do that with a standard user-defined fitting function, it is only moderately difficult with an all-at-once fitting function.

To learn more about all-at-once fitting functions, execute this command on Igor's command line: DisplayHelpTopic "All-At-Once Fitting Functions"

April 19, 2019 at 09:21 am - Permalink