Probabilistic Methods in
Engineering
Lecture 12
Dr. Maha Ali
r
• 2 or more Random
variables.
• Correlation and
correlation Coefficient
• Continuous Random
Variable
PMF of a R.V. quick review
Px X
rolling 2 fair 4 sided dice
is the maximum number
11 2 314 1 2 2,1
2121
2
00 2 3
Px
Experiment and 2 R.V.
Sometimes it is helpful; to look at it
as a vector geometry in 2-D
As a point in the
XY plane
Model as 2 d R.V
Location on GPS
High and low temperature of the day
Joint Random Variables
Joint PMF
it
i U
2,4 0 X 217 0
PxCX Pay x y
Pyls Px y 451
iClicker
The joint PMF of the random variables X and Y is given by the
following table:
1,4 1
C C 2Ce 20 0 4 3C
a) Find the value of the constant c. C 6C 20 C
b) Find pY (2).
C
Py 2 2 0 0 42 6C
Conditioning one Random Variable on another
• Let X and Y be two discrete random variables. The conditional PMF of
X given Y is
III
P X Y PC
p
Conditional pmf cont.
is 1
Pxyy
a
Conditional Probability Cont.
x Y Py
i
Px y x b Px y b
Px y 451 Px y X YR Px X
iClicker PCA B P
[Link] t
9 1 PY 1 4 0
P aE
iClicker
Example
Consider a transmitter that is sending messages over a computer network. Let us define the following two
random variables:
X : the travel time of a given message,
Y: the length of the given message.
The length of the message is 102 bytes with probability 5/6 and 104 bytes with Probability 1/6.
We also know the PMF of the travel time of a message that has a given length :
i i
We want to find the (unconditional) PMF of the travel time of a message.
7 4 FRIED Pay
Px 182 F Px 10 f 21100 t t
Independence of Random Variables P AAB PA PI
Px y x s
Px Py y for all x y
Pay Ny Px x
for ally with Pyly o and alla
a
PINYI
PUI Patty
Conditional Independence
There is a similar notion of conditional independence of two random variables, given
an event A with P(A) > o. The conditioning event A defines a new universe and all
probabilities (or PMFs) have to be replaced by their conditional counterparts. For
example, X and Y are said to be conditionally independent, given a positive probability
event A, if
P X x Y 1 A P X X A PLY A
for all x y
Pxy 5 Px a x Pya's for all x y
Pxly A X B Px A 41
Expectation of joint independent Random
Variables
If X and Y are independent random variables, then
E E X E Y
[Link] 4
XY 14
II
X Px x Y Py 41
EEX E Y
E g X h Y E 94 E KEY
Variance of Sum of two or more independent
Random Variables
Var Y Var X Var Y
proof do on your own
Var Xn Xz Xst Xn Var a Varcalt
an
Example
. The joint PMF of the random variables X and Y is given by the
following table:
Z YX
i
E Z 4 2 2 2 44 2
c) Consider the random variable Z = Y X2. Find E[Z | Y = 2].
zE 2
2 2
PXlyLN2 2E 414 23
1
from part b 0 5 f
Pala E py 2
if 2
3
EE EE Y 1;2 EXPayCN2II
Joint Moments, Correlation, and Covariance
• Kth moment is E
• Kth central moment is E X M
• Second central moment = Variance = E X µF
• For 2 R.V we can raise each to a different power and we get:
n
• ( j,k)th moment = E x y
• ( j,k)th central moment = E ex MY CY Ms
Special cases
• j=k=1
E XY the correlation of 24
Cov X Y E CX Mx LY MB E XY M My
covariance of X and y
can be we a ur
covariance
of against itself E X2 M VAR X
alohg positive
Correlation Coefficient
normalize the covariance to measure the correlation in an absolute scale.
The correlation coefficient of X and Y is defined as:
FARM
1 2 12 1
Scatter plot: Perform experiment and observe X and Y
MY
DY
I
f
I
o
Correlation coefficient =1
D
Li f 1
Can
knowing get Y