Covariance is a measure of how changes in one variable are associated with changes in a second variable. It’s a measure of how they Co (together) Vary (move) or how they move in relation to each other.

Interpreting Covariance:

A large positive or negative covariance indicates a strong relationship between two variables. However, you can’t necessarily compare covariances between sets of variables that have a different scale, since the covariance of variables that take on high values will always be higher since covariance values are unbounded, they could take on arbitrarily high or low values. This means that you can’t compare the covariances between variables that have a different scale. A positive covariance variable that has a large scale will always have a higher covariance than a variable with an equally strong relationship, yet smaller scale.

Covariance is a measure of the joint variability of two random variables. It is essentially measured as positive, negative, or neutral (i.e. no covariance). If you take two random variables, and the larger numbers of one, tend to corresopond with the larger numbers of the other, then there is a positive covariance. If the opposite holds true, then the covariance would be negative. If there one variable does not correspond at all with the other, you would have no covariance.

Last Updated: April 07, 2019