Suppose I have n=3 of ranked ordered list of items:

R1=<i1, i2, i3>
R2=<i2,i1, i3>
R3=<i1, i2, i3>

I want to compute the correlation of an ordered list R= of same items.

1) Is "Kendall rank" the right correlation coefficient to use?

2) If yes, how to calculate the correlation average between R and other lists Ri since the coefficient returns negative values. Should I take negative values, though?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.