2
$\begingroup$

Lets say I am given the following table that shows the joint probability function of X and Y:

$$\begin{array} \\{}&y=1&y=2&y=3 \\x_=1&0.1&0.2&0.1 \\x=2&0.1&0.25&0.15 \\x=3&0&0&0.1 \end{array}$$

Then I am fairly sure the marginal probability functions of $X$ and $Y$ are given by their totals across each row and column, eg:

$$\begin{array} \\{}&y=1&y=2&y=3&P_X(X) \\x=1&0.1&0.2&0.1&0.4 \\x=2&0.1&0.25&0.15&0.5 \\x=3&0&0&0.1&0.1 \\P_Y(Y)&0.2&0.45&0.35&1 \end{array}$$

I want to workout $$E(X),E(Y),Var(X),Var(Y),Cov(X,Y)$$

Now I assume that E(X) comes from P(X) here, so I get $ 0.4*1 + 0.5 * 2 + 0.1 * 3=1.7$ E(Y) same procedure $= 2.15$

Var(X) = $ 0.4*1^2 + 0.5 * 2^2 + 0.1 * 3^2 - 1.7^2 = .41$ Var(Y) same procedure = .5275

Cov(X,Y) I am not sure how to calculate this really. Is any of the above wrong? Any hints on solving Cov(X,Y)? Thank you for your time, and I am sorry about any bad latex formatting, this much latex actually slows my little laptop to a crawl.

$\endgroup$
1
$\begingroup$

The covariance of $X$ and $Y$ is $E(XY)-E(X)E(Y)$. (The formal definition of covariance is $E((X-E(X))(Y-E(Y)))$, but that is usually, and in this case, harder to work with.)

To find $E(XY)$, find the sum $\sum_{(x,y)} xy\Pr(X=x\land Y=y)$.

There will be $9$ terms to add up, really only $7$, since $2$ of the terms are $0$,

A typical term like the one for $x=2$, $y=3$ in your table makes a contribution of $(2)(3)(0.15)$ to $E(XY)$.

Remark: The procedure of your other calculations is correct. I have not checked the arithmetic.

$\endgroup$
  • $\begingroup$ Thank you once again André, that was very clear. $\endgroup$ – Display Name Apr 17 '14 at 6:08
  • $\begingroup$ You are welcome. $\endgroup$ – André Nicolas Apr 17 '14 at 6:09
  • $\begingroup$ Related query, if I were to have a function $G = 10 + 10X + 15Y$ using the above $X$ and $Y$, how would I go about working out the mean and standard deviation of $G$? Does the mean of G just multiply through the Mean of $X$ and $Y$ in place of X and Y I the equation? Same with taking the square root of variance and plugging it in? $\endgroup$ – Display Name Apr 17 '14 at 6:12
  • $\begingroup$ The mean of $G$ is $10+10E(X)+15E(Y)$. Means are linear. Variance is more complicated. The variance of $a+bX+cY$ is $b^2\text{Var}(X)+c^2\text{Var}(Y)+2bc\text{Cov}(X,Y)$. If $X$ and $Y$ are uncorrelated (covariance is $0$, which happens when $X$ and $Y$ are independent, and in some other cases) the covariance term is $0$, which is nice. Not in your example! $\endgroup$ – André Nicolas Apr 17 '14 at 6:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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