The joint probability - two coins

6 replies · 3K views · last reply
docnet
Messages
796
Reaction score
487
Homework Statement
.
Relevant Equations
.
Typed problem 7 about two coins: an unbiased coin 1 and a biased coin 2 with P(H) = 0.75, defining X and Y from tosses, with parts (a) to (d) asking for the joint PMF, marginals, E(Y) and E((X-0.5)Y).
(a)

The probability mass function of X and Y:

$$p_{X,Y}(1,1)= \frac{1}{2}\cdot \frac{3}{4}=\frac{3}{8}$$
$$p_{X,Y}(1,0)=\frac{1}{2}\cdot \frac{1}{4}=\frac{1}{8}$$
$$p_{X,Y}(0,1)= \frac{1}{2}\cdot \frac{3}{4}=\frac{3}{8}$$
$$p_{X,Y}(0,0)= \frac{1}{2}\cdot \frac{1}{4}=\frac{1}{8}$$

or

$$
p_{X,Y}(a,b)=
\begin{cases}
\frac{3}{8} & (a,b)=(1,1) \quad \text{or}\quad(a,b)=(0,1) \\
\frac{1}{8} & (a,b)=(1,0) \quad \text{or}\quad(a,b)=(0,0)
\end{cases}$$

(b)

The marginal distribution of X is

$$p_X(1)=p_{X,Y}(1,0)+p_{X,Y}(1,1)=\frac{1}{2}\frac{3}{4}+\frac{1}{2}\frac{1}{4}=\frac{1}{2}$$
$$p_X(0)=p_{X,Y}(0,0)+p_{X,Y}(0,1)=\frac{1}{2}\frac{1}{4}+\frac{1}{2}\frac{3}{4}=\frac{1}{2}$$

The marginal distribution of Y is

$$p_Y(1)=p_{X,Y}(0,1)+p_{X,Y}(1,1)=\frac{1}{2}\frac{3}{4}+\frac{1}{2}\frac{1}{4}=\frac{3}{4}$$
$$p_Y(0)=p_{X,Y}(0,0)+p_{X,Y}(1,0)=\frac{1}{2}\frac{1}{4}+\frac{1}{2}\frac{3}{4}=\frac{1}{4}$$

(c)

The expected value of Y is

$$E(Y)=1\cdot p_Y(1)+0\cdot p_Y(0)=\frac{3}{4}$$

(d)

$$E((X-0.5)\times Y)=\sum\limits_{(a,b)}(a-.5)bp_{X,Y}(a,b)$$
$$(1-0.5)(1)p_{X,Y}(1,1)+(1-0.5)(0)p_{X,Y}(1,0)+$$
$$(0-0.5)(1)p_{X,Y}(0,1)+(0-0.5)(0)p_{X,Y}(0,0)=$$
$$\frac{1}{2}\cdot\frac{3}{8}+0-\frac{1}{2}\cdot \frac{3}{8}+0=0$$
 
docnet said:
Homework Statement:: .
Relevant Equations:: .

View attachment 293161(a)

The probability mass function of X and Y:

$$p_{X,Y}(1,1)= \frac{1}{2}\cdot \frac{3}{4}=\frac{3}{8}$$
$$p_{X,Y}(1,0)=\frac{1}{2}\cdot \frac{1}{4}=\frac{1}{8}$$
$$p_{X,Y}(0,1)= \frac{1}{2}\cdot \frac{3}{4}=\frac{3}{8}$$
$$p_{X,Y}(0,0)= \frac{1}{2}\cdot \frac{1}{4}=\frac{1}{8}$$
That's not what I get. If ##X = 1##, then we toss the unbiased coin (coin 1) again.
 
PeroK said:
That's not what I get. If ##X = 1##, then we toss the unbiased coin (coin 1) again.

yes! the problem I solved isn't same as the problem in the prompt :(

but don't worry, i think i know what the problem is asking, and how to solve it
 
OK so here's what I have

(a)

$$P_{X,Y}(1,1)=\frac{1}{2}\frac{1}{2}=\frac{1}{4}$$
$$P_{X,Y}(1,0)=\frac{1}{2}\frac{1}{2}=\frac{1}{4}$$
$$P_{X,Y}(0,1)=\frac{1}{2}\frac{3}{4}=\frac{3}{8}$$
$$P_{X,Y}(1,1)=\frac{1}{2}\frac{1}{4}=\frac{1}{8}$$(b)

$$P_X(1)=\frac{1}{4}+\frac{1}{4}=\frac{1}{2}$$
$$P_X(0)=\frac{3}{8}+\frac{1}{8}=\frac{1}{2}$$
$$P_Y(1)=\frac{3}{8}+\frac{1}{4}=\frac{5}{8}$$
$$P_Y(0)=\frac{1}{8}+\frac{1}{4}=\frac{3}{8}$$

(c)

$$E(Y)=\frac{5}{8}+0\cdot \frac{3}{8}=\frac{5}{8}$$

(d)

$$E((X-.5)\times Y)=(\frac{1}{2}\frac{1}{2}-\frac{1}{2}\frac{1}{2})\frac{5}{8}=0$$
 
It looks right apart from the answer to d). Note that in general ##E(XY) \neq E(X)E(Y)##. (Another simple but erroneous calculation on your part!) But, in general, ##E(X + Y) = E(X) + E(Y)##. That gives you two ways to do part d): go through the four options, or use the expected value of a sum.
 
@PeroK thank you :) is this what you mean?

Displayed formula: E[g(X, Y)] equals the double sum over (x, y) of g(x, y) p(x, y), the expectation of a function of two discrete random variables.

using the formula:
$$E((X-.5)Y)=\frac{1}{8}-\frac{3}{16}=-\frac{1}{16}$$ 😒using linearity and the definition of covariance:

$$E((X-.5)Y)=E(XY)-\frac{5}{16}$$
Formula for covariance of discrete random variables X and Y: Cov(X,Y) equals the double sum over (x,y) in S of (x minus mu_X)(y minus mu_Y) times f(x,y).

$$E(XY)=cov(X,Y)+E(X)E(Y)=-\frac{1}{16}+\frac{5}{16}$$

$$E((X-.5) Y)=-\frac{1}{16}$$ 😇
 
Attachments
  • Formula for covariance: cov(X, Y) equals E(XY) minus the product of the means mu_x and mu_y.
    Screen Shot 2021-11-28 at 11.24.17 AM.png
    2.8 KB · Views: 224
Last edited:
  • Like
Likes   Reactions: PeroK
Yes, it's good to see you do it both ways and get the same answer!

Note: the quick and valid way to get ##E(XY)## is to note that ##XY## is only non-zero when ##X = Y = 1##, which happens with probability ##1/4##. Hence ##E(XY) = \frac 1 4##.
 
  • Love
Likes   Reactions: docnet