How do I perform nonlinear regression

I have 60 data pairs. Each data pair include 3 inputs and 1 output. The trends (2 dimension) between output and individual inputs are complex, which are not easily presented by a formula. I want to use nonlinear regression to find the relationship between inputs and output.
For nonlinear regression, normally I can use "mdl = fitnlm(tbl,modelfun,beta0)" to do regression. The fitnlm attempts to find values of the parameters in modelfun.
My question is: if I don't know the formula of modelfun, how can I use fitnlm.
Are there any other methods to be used?
Thank you.

回答(0 个)

类别

标签

提问:

2015-3-23

编辑:

2015-3-23

Community Treasure Hunt

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

Start Hunting!

Translated by