Problems with Regress Function

4 views (last 30 days)
Robert
Robert on 5 Nov 2014
Commented: the cyclist on 5 Nov 2014
I'm trying to make a linear regression I have a matrix 1439x5. The first column is observed data and the following 4 are predictor variables. my code looks like this
ans = regress(wl(:,1:1),wl(:,2:5))
I am only getting an output of 4 numbers, where there should be 1439. Any ideas?
Thank you in advance for any help

Accepted Answer

the cyclist
the cyclist on 5 Nov 2014
You say that you have 4 predictor variables. Therefore, you are estimating the coefficients of those 4 predictive terms in the linear regression. You get exactly what I would expect out of the function.
Why do you expect something else? Have you read
doc regress
?
  2 Comments
Robert
Robert on 5 Nov 2014
Hi, you're too right. For some reason I thought it was going to spit out 1439 predicted values based on the coefficients, but re reading the regress doc made me realise I am wrong. Cheers
the cyclist
the cyclist on 5 Nov 2014
Glad it worked out. FYI, you can also use fitlm to fit linear regression models. This is the more object-oriented version that MATLAB introduced recently, and has a bunch of methods (such as predict) that can do things like get the predicted values.

Sign in to comment.

More Answers (0)

Community Treasure Hunt

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

Start Hunting!