Correlation between several variables

67 views (last 30 days)
Could anyone help me with the code please, please?
I have a table of 11 variables. Each variable has 500 observations.
The first ten variables (x1 through to x10) are the so-called predicotrs , the last variable is y, which is the dependent variable.
How can I compute the correlation of y with each of the 10 predictors and also determine their signidicance?
Is there a command that allows me to present the result in some sort of table?
Or do I have to construct the for loop?
The task is to find the most 3 correlated variables with y and the least 3 correlated varaibles with y.
My code:
head(Data)
for i=1:10
vari=Data(:,i)
var11=Data(:,11)
corr(vari,var1)
end
Do you think this can be improved?
Why can I not do corr(Data(:,1:10),Data(:,11))?

Accepted Answer

the cyclist
the cyclist on 23 Apr 2021
Edited: the cyclist on 23 Apr 2021
I think I would use the corrcoef function:
[r,p] = corrcoef(Data);
and just focus on the last column of the r and p outputs, which will be the correlation and p-value of the y with the 10 explanatory variables.

More Answers (0)

Categories

Find more on Matrices and Arrays in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!