How do I compute the mean correlation from a n by n correlation matrix?

I have a 34 x 34 correlation matrix and would like to compute the mean correlation (i.e., average of correlations above or below the diagonal). I was wondering if anyone could show me how to do that.

Answer 1

[Jyotish Robin]

Hi Liane!

You could accomplish the task by making use of the functions 'tril' and 'mean'.

If a is your matrix, the following command returns the upper triangular elements (not including the diagonal elements) as a column vector:

colvect= a(find(~tril(ones(size(a)))))

Now if you do

mean(colvect)

It will give you the average of upper triangular entries of the matrix a.

In a similar way, you could deal with the lower triangular entries also. You could learn more about 'tril' and 'mean' functions from the following documentations:

• https://www.mathworks.com/help/matlab/ref/tril.html
• https://www.mathworks.com/help/matlab/ref/mean.html

Hope this helps!

Regards,

Jyotish