How to calculate regression with dummy variables?

10 views (last 30 days)
Hi, I am trying to use 2nd degree polynomial regression with dummy variables. Which function should I use? Thanks for any help!

Accepted Answer

the cyclist
the cyclist on 17 Jul 2018
Edited: the cyclist on 17 Jul 2018
If you have the Statistics and Machine Learning Toolbox, take a look at the fitlm function.
You might also find the dummyvar command useful for creating dummy variables.
Take a look at Regression with Categorical Covariates for an example.
  2 Comments
JFz
JFz on 17 Jul 2018
Thanks! But fitlm is linear model. I need polynomial (2 degree or 3 degree). What function should I use? Should I use fitnlm or polyfit? Thanks again! Jennifer
the cyclist
the cyclist on 17 Jul 2018
Edited: the cyclist on 17 Jul 2018
Do you mean that you have polynomials for your independent variables, like
y = a + b*x + c*x^2
That is a linear model. "Linear" refers to the coefficients, not the powers (or whatever) of the variables.
On the other hand, something like
y = a + x^b
is nonlinear in the coefficient. Yes, you would need fitnlm for something like that.

Sign in to comment.

More Answers (0)

Tags

Products


Release

R2018a

Community Treasure Hunt

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

Start Hunting!