Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Given two uncorrelated random variables $X,Y$ with the same variance $\sigma^2 $ I need to compute $\rho= \frac{COV(X,Y)}{\sigma(X)\sigma(Y)}$ between $X+Y$ and $2X+2Y$. I know it should be a number between $-1$ and $1$ and I don't understand how come I get $4$.

Here's what I did:

$COV(X+Y,2X+2Y)=COV(X+Y,2X)+COV(X+Y,2Y)=COV(2X,X)+COV(2X,Y)+COV(2Y,Y)+COV(2Y,X)=2COV(X,X)+2COV(Y,Y)+4COV(X,Y)=2\sigma^2+2\sigma^2=4\sigma^2$ so final result is $\rho=4$ since $\sigma(X)=\sqrt{Var(x)}$.

What's wrong with what I did?

share|cite|improve this question
up vote 4 down vote accepted

The reason things went wrong is probably due to an unfortunate choice of notation, the use of $X$ and $Y$ with two different meanings.

We want the correlation coefficient $\rho(U,V)$, where $U=X+Y$ and $V=2(X+Y)$. So we need to divide $\text{Cov}(U,V)$ by the product of the standard deviations of $U$ and of $V$ (not of $X$ and of $Y$).

You did divide, but by the wrong thing. For the denominator, calculate and use $\sigma_U\sigma_V$.

share|cite|improve this answer
@Ben Benli: What I did was to explain what went wrong in your calculation. Please note that the answer by did gives a much smoother path to the answer. – André Nicolas Oct 25 '12 at 19:48

Show that, for every nondegenerate random variable $Z$ and nonzero real number $a$, $\mathrm{var}(aZ)=a^2\cdot\mathrm{var}(Z)$ and $\mathrm{cov}(Z,aZ)=a\cdot\mathrm{var}(Z)$. Deduce that $\varrho(Z,aZ)=\mathrm{sgn}(a)$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.