Covariance in a joint probability mass function

2.6k Views Asked by At

I'm trying to solve this question but I'm still a little iffy on joint distribution functions:

Let $X$ and $Y$ be two independent discrete random variables with the following Joint probability mass function

$Y = 0$ $Y = 1$
$X = 0$ $\frac{1}{2}$ $\frac{1}{4}$
$X = 1$ $\frac{1}{4}$ 0

Find $\operatorname{Cov}(5X - \frac{1}{2}, Y + 4)$.

Any help on this?

I tried a few ways but none have really worked. I tried $\operatorname{Cov}(XY) = E(XY) - E(X)E(Y)$, but couldn't figure out how to fit it here. Same for the method where I have to find the mean of each. I'd really appreciate any help, would go a long way!

1

There are 1 best solutions below

2
On BEST ANSWER

Indeed: $$\mathsf{Cov}(X,Y)=\mathsf E(XY)-\mathsf E(X)\,\mathsf E(Y)$$

And, since $X$ and $Y$ are Bernoulli: $$\begin{align}\mathsf E(XY)&=\sum_{x,y} xy\,\mathsf P(X\,{=}\,x,Y\,{=}\,y)\\[1ex]&=\mathsf P(X\,{=}\,1,Y\,{=}\,1)&&xy=0\text{ when either is }0\\[2ex]\mathsf E(X)&=\sum_{x,y} x\,\mathsf P(X\,{=}\,x,Y\,{=}\,y)\\[1ex]&=\mathsf P(X\,{=}\,1,Y\,{=}\,0)+\mathsf P(X\,{=}\,1,Y\,{=}\,1)\\[2ex]\mathsf E(Y)&=\sum_{x,y}y\,\mathsf P(X\,{=}\,x,Y\,{=}\,y)\\[1ex]&=\mathsf P(X\,{=}\,0,Y\,{=}\,1)+\mathsf P(X\,{=}\,1,Y\,{=}\,1)\end{align}$$

By the Bilinearity of Covariance when $a,b,c,d$ are constants and $W,X,Y,Z$ are random variables:

$$\mathsf{Cov}(aX+bW,cY+dZ)={{ac\,\mathsf{Cov}(X,Y)}+{ad\,\mathsf{Cov}(X,Z)}\\+{bc\,\mathsf{Cov}(W,Y)}+{bd\,\mathsf{Cov}(W,Z)}}$$

In this case:

$$\mathsf{Cov}(aX+b, cY+d)=ac\,\mathsf{Cov}(X,Y)$$