Covariance of Linear Combination of Random Variables with Another
Jump to navigation
Jump to search
Theorem
Let $X, Y, Z$ be random variables.
Let $a, b$ be real numbers.
Then:
- $\cov {a X + b Y, Z} = a \cov {X, Z} + b \cov {Y, Z}$
Proof
\(\ds \cov {a X + b Y, Z}\) | \(=\) | \(\ds \expect {\paren {a X + b Y} Z} - \expect {a X + b Y} \expect Z\) | Covariance as Expectation of Product minus Product of Expectations | |||||||||||
\(\ds \) | \(=\) | \(\ds a \expect {X Z} + b \expect {Y Z} - \paren {a \expect X + b \expect Y} \expect Z\) | Expectation is Linear | |||||||||||
\(\ds \) | \(=\) | \(\ds a \paren {\expect {X Z} - \expect X \expect Z} + b \paren {\expect {Y Z} - \expect Y \expect Z}\) | ||||||||||||
\(\ds \) | \(=\) | \(\ds a \cov {X, Z} + b \cov {Y, Z}\) | Covariance as Expectation of Product minus Product of Expectations |
$\blacksquare$