Skip to content
Formula

Covariance and variance of and

Statistics • Correlation and Regression

Statement

For variates , : and . If and are uncorrelated, , so .

How to use it

  1. 1Why it works: covariance is the mean of the product of deviations from the means. The deviation of is , and that of is . Multiply the brackets and take the mean of each term: , , .
  2. 2So gives , and gives the variances.
  3. 3Uncorrelated means : drop the middle term.

Worked example

, and . Find and .

Step-by-step

  1. 1

    Why: Square of the bracket .

  2. 2

    Why: : the cross terms cancel.

Answer

and

Watch out

adds the two variances; it does not subtract them. Only the cross term changes sign.

Builds on

Also known as

variance of sum of uncorrelated variables, covariance of x+y and x-y, correlation between x+y and x-y

Appears in past papers

2016 (II)