Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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

there's a something that I though about that Disturbing me for quite some time now:

Let $X$ , $Y $ be a random variables. I call random variables independent if for every possible value of $x_{i}$ that $X$ might get and $y_{j}$ that $Y$ might get:

$p(X=x_{i} \cap Y=y_{j})=P(X=x_{i})\cdot P(Y=y_{j})$.

In addition two random variables are Non-correlated if $ E[X\cdot Y]=E[X]\cdot E[Y]$,

when $E$ represents the expected value.

I know that if $X$ and $Y$ are independent so they are Non-correlated.

when $X$ and $Y$ are dependent, I can think of only cases which $E[X\cdot Y]=0$ because one of the expected value of $X$ or $Y$ is o.

my QUESTION is: Are they any dependent random varaibles and non-correlated so that $E[X]\neq0$ and $E[Y]\neq0$, and if not, how does one prove that?

Thank you for the help.

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

Yes. Just take your favourite example of $X, Y$ dependent and uncorrelated with expectation 0, and consider the new variables $X+1$ and $Y+1$. They're still dependent and uncorrelated, but now they have expectation 1.

Explicitly: a simple example with $E[X]=E[Y]=0$ is for $(X,Y)$ to be uniformly distributed in the set $\{(0,1),(1,0),(0,-1),(-1,0)\}$. Adding $1$ to $X$ and $Y$, we obtain new random variables $U=X+1$ and $V=Y+1$ such that $(U,V)$ is uniformly distributed in $\{(1,2),(2,1),(1,0),(0,1)\}$.

We have $E[U]=\frac{1+2+1+0}{4}=1$, $E[V]=\frac{2+1+0+1}{4}=1$, $E[UV]=\frac{2+2+0+0}{4}=1=E[U]E[V]$, so $U$ and $V$ are uncorrellated. But $P(U=1 \cap V=1)=0 \neq \frac{1}{2} \cdot \frac{1}{2}=P(U=1)P(V=1)$, so $U$ and $V$ are dependent.

share|cite|improve this answer
Do you mean by $X+1$ to add to every $x_{i}$ $1$? – henry11 Apr 6 '11 at 18:02
@henry11: That is what he means: you get $Pr(X+1 = x_i +1) = Pr(X = x_i)$. Note that two random variables are uncorrelated if their covariance $E[(X-\mu_X)(Y-\mu_Y)] = 0$, and that adding 1 to $X$ and to $Y$ also adds 1 to their means so leaving the covariance unchanged. – Henry Apr 6 '11 at 18:41
I'll need much more simple explanation, I don't know what is covariance yet, and what the expression you wrote means. My level is just up to what I wrote. – henry11 Apr 6 '11 at 18:44

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.