CFA Level 1 – Quantitative Methods, Correlation and Covariance
Covariance is a measure of how two data series change together. It can be a difficult number to interpret since it is not in percentage terms or units of either data series. For this reason Covariance is not frequently used in isolation but as a basis for other statistical measures. The most common of these is Correlation.
The Coefficient of Correlation (often referred to simply as the Correlation) is the Covariance divided by the product of the two standard deviations for both series. This results in a number between -1 and 1, with 1 indication a perfect linear relationship between the two series and -1 indicating a perfect negative relationship between the series.
It should be noted that these measures are entirely linear and do not account for any non-linearities in the relationship.