Sanakirja.org

(statistics) A statistical measure defined as ${\displaystyle \scriptstyle \operatorname {Cov} (X,Y)=\operatorname {E} ((X-\mu )(Y-\nu ))}$ given two real-valued random variables X and Y, with expected values ${\displaystyle \scriptstyle E(X)\,=\,\mu }$ and ${\displaystyle \scriptstyle E(Y)\,=\,\nu }$.

The sign of the covariance of two random variables X and Y

(statistics) A statistical measure defined as ${\displaystyle \scriptstyle \operatorname {Cov} (X,Y)=\operatorname {E} ((X-\mu )(Y-\nu ))}$ given two real-valued random variables X and Y, with expected values ${\displaystyle \scriptstyle E(X)\,=\,\mu }$ and ${\displaystyle \scriptstyle E(Y)\,=\,\nu }$.

Geometric interpretation of the covariance example. Each cuboid is the axis-aligned bounding box of its point (x, y, f (x, y)), and the X and Y means (magenta point). The covariance is the sum of the volumes of the cuboids in the 1st and 3rd quadrants (red) and in the 2nd and 4th (blue).