Correlation between multiple columns of table

Hi guys!

I have a table (see attached) for which I want to calculate corrcoef ([r p]) between multiple columns. Eg. between column 2 and column 6, column 2 and column 7, then column 3 and column 6, column 3 and column 7. The last combination would be column 5 and column 12.

Is there a way to do it a with a loop (or any other way, as a matter of fact) and automatically store the values in a new table so that I don't mix them up?

Thanks in advance!

I'm on Matlab 2019a.

Answer 1

[KSSV]

iwant = zeros(2,2,[]) ;
count = 0 ; 
for r = 2:3
    for p = 6:7
        count = count+1 ; 
        iwant(:,:,count) = corrcoef(data(:,r),data(:,p)) ; 
    end
end
iwant(:,:,count+1) = corrcoef(data(:,5),data(:,12)) ; 

[Daphne PARLIARI]

Thank you for your answer.

I tried your way but I get

Error using corrcoef (line 114)

Subscripting a table using linear indexing (one subscript) or multidimensional indexing (three or more subscripts) is

not supported. Use a row subscript and a variable subscript.

[KSSV]

How did you read the file? What I have shown in the code is for a matrix. If you have a table, the indexing differs. You need to use data.(i) etc.

[Daphne PARLIARI]

Actually the attached file is a double that I have produced earlier in the code, it's the outerjoin between two tables.