Abstract
Suppose we have two random variables that have given probability distributions. Is it possible that their correlation coefficient can assume any value in [ − 1 , 1 ] ? It turns out that the answer ...

This publication has 6 references indexed in Scilit: