Create a new equation from a model in order to nonlinear, polynomial multiple regression, etc. for multiple many (4 or above) variables

1 次查看(过去 30 天)
Hello. There are column vectors of at least 1000 data of 6 different types (variables). These parameters, called "q, E, w, x, σ, KP", represent each column vector and each variable. Using these column vectors, I want to suggest a new equation via nonlinear, polynomial multiple regression, etc. methods for multiple (4 or more) variables. What is the way to create a new equation from a model for? In the MATLAB Curve Fitting toolbox, I saw that a model can be created for, "X, Y, Z" inputs, a maximum of 3 variables. What's the way to can be done that?
My aim: Create an equation: f(q, E, w, x, σ, KP)=y and show this mathematically model like "q*a1+E*a2+w*a3+w*a4+x*a5+σ*a6+KP*a7" or different degree, variation and mathematical approaching.

回答(1 个)

Abdolkarim Mohammadi
Your have a linear relationship and the best option for you is the multiple linear regression. See the documentation of the regress() function:
https://www.mathworks.com/help/stats/regress.html

类别

Help CenterFile Exchange 中查找有关 Linear and Nonlinear Regression 的更多信息

产品


版本

R2020a

Community Treasure Hunt

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

Start Hunting!

Translated by