STAT422 - Slides

STAT 422: Applied Probability Lecture Notes
Fastel Chipepa1
Botswana International University of Science and Technology1
February 7, 2024
F. Chipepa February 7, 2024 1 / 51

Outline
1 CHAPTER 1
1.1: Sample Space and Events
1.2: Probabilities Defined on Events
1.3: Conditional Probability
1.4: Independent Events
1.5: Bayes’ Formula
2 CHAPTER 2
2.1: Random Variables
2.2: Discrete Random Variables
2.3: Continuous Random Variables
2.4: Expectation of a Random Variable
2.5: Jointly Distributed Random Variables

CHAPTER 1 1.1: Sample Space and Events
Sample Space and Event
Sample Space (S) - is the set of all possible outcomes

e.g If you flip a coin S={H, T}
If you roll a die S={1, 2, 3, 4, 5, 6} .
The set(a, b) is defined to consist of all points x such that a x b, [a,
b] consist of all points x such that a ¤ x ¤ b, (a, b] consist of all points x
such that a x ¤ b and [a, b) consist of all points x such that a ¤ x b.
Event - is any subset of the sample space.
For any two events E and F of a sample space S, we define the new event
E Y F to consist of all outcomes that are either in E or F or in both E
and F.
E X F consist of all outcomes which are both in E and F. If E X F H,
then E and F are mutually exclusive.

CHAPTER 1 1.1: Sample Space and Events
8
n1 En , is defined to be the event that consists of all outcomes that are
in En for at least one value of n 1, 2, 3, ....
8
n1 En , is defined to be the event consisting of those outcomes that are
in all of the events En , n 1, 2, 3, ....
For an event E, E c (compliment of E) consists of all outcomes in the
sample space S that are not in E.

CHAPTER 1 1.2: Probabilities Defined on Events
Probabilities Defined on Events
Given a sample space S, for each event E, we assume that a number P(E)
is defined and satisfies the following 3 conditions
i 0 ¤ P pE q ¤ 1
ii P pS q 1
iii For any sequence of events E1 , E2 , ..., that are mutually exclusive, then
pp 8
°8
n1 En q n1 P pEn q, where P(E) is the probability of event E.

Since the events E and E c are always mutually exclusive and since
E Y E c S, we have by (ii) and (iii) that
1 P pS q P pE Y E c q P pE q P pE c q or P pE c q 1 P pE q
For events E and F P pE q P pF q P pE Y F q P pEF q or
P pE Y F q P pE q P pF q P pEF q, where EF E X F .
P pE Y F q P pE q P pF q if E and F are mutually exclusive.

Given 3 events E, F, and G

P pE Y F Y Gq P ppE Y F qGq which is equivalent to
P pE Y F q P pGq P ppE Y F qGq.
P pE Y F Y Gq P pE q P pF q P pEF q P pGq P pEG Y F Gq
P pE q P pF q P pEF q P pGq P pEGq P pF Gq P pEGF Gq
P pE q P pF q P pGq P pEF q P pEGq P pF Gq P pEF Gq
Which can be shown by induction
° for n°events E1 , E2 , ..., En that
P pE1 Y E2 Y ... Y En q i1 P pEi q i j P pEi Ej q
° °
i j k P pEi Ej Ek q i j k l P pEi Ej Ek El q ... p1q P pE1 E2 ...En q.
n 1

Homework 1
A drug has the following information:

a) There is a 10% chance of experiencing headache (H).
b) There is a 15% chance of experiencing nausea (N).
c) There is a 5% chance of experiencing both side effects.
What is the probability of experiencing at least one side effect?

CHAPTER 1 1.3: Conditional Probability
Conditional Probability
Often we want to calculate the probability that an event E occurs given

that an event F occurs. We use the notation P pE |F q. This defined when
P pF q ¡ 0.
The rule for conditional probability
P pE |F q P pPEpF
XF q
q
Example 1.1: Suppose cards numbered one through ten are placed in a
hat, mixed up, and then one of the cards is drawn. If we are told that the
number on the drawn card is at least five, then what is the conditional
probability that it is ten?
Solution: Let E denote the event that the number of the drawn card is
ten, and let F be the event that it is at least five. The desired probability
id P pE |F q. However, P pE X F q E since the number of the card will be
both ten and at least five if and only if it is the number 10. Hence,
P pE |F q 106 6.
1
1
10

CHAPTER 1 1.3: Conditional Probability
Homework 2
1. Suppose an urn contains seven black balls and five white balls. We draw
two balls from the urn without replacement. Assuming that each ball in
the urn is equally likely to be drawn, what is the probability that both
drawn balls are black?
2. Suppose that each of three men at a party throws his hat into the center
of the room. The hats are first mixed up and then each man randomly
selects a hat. What is the probability that none of the three men selects
his own hat?

CHAPTER 1 1.4: Independent Events
Independent Events
Two events are said to be independent if

P pE X F q P pE qP pF q.
Implications:
P pE |F q P pE q and P pF |E q P pF q.
Pairwise independence does not necessarily imply joint independence.
Example 1.2: Let a ball be drawn from an urn containing four balls,
numbered 1, 2, 3, 4. Let E t1, 2u, F t1, 3u, G t1, 4u. If all four
outcomes are assumed equally likely, then
P pE X F q P pE qP pF q 14
P pE X G P pE qP pGq 14
P pF X Gq P pF qP pGq 41 .
However, 14 P pE X F s X Gq P pE qP pF qP pGq.
Hence, even though the events E, F, G are pairwise independent, they are
not jointly independent.

CHAPTER 1 1.5: Bayes’ Formula
Bayes’ Formula
- Let E and F be events.

We may express E pE X F qYpE X F c q because in order for a point to be
in E, it must either be in both E and F, or it must be in E and not in F.
Since pE X F q and pE X F c q are mutually exclusive, we have that
P pE q P pE X F q P pE X F c q
P pE |F qP pF q P pE |F c qP pF c q
P pE |F qP pF q P pE |F c qp1 P pF qq
The result state that the probability of event E is a weighted average of

the conditional probability of E given that F has occurred and the
conditional probability of E given that F has not occurred, each
conditional probability being given as much weight as the event on which
it is conditioned has of occurring.

Example 1.3: Consider two urns. The first contains two white and seven
black balls, and the second contains five white and six black balls. We flip
a fair coin and then draw a ball from the first urn or the second urn
depending on whether the outcome was heads or tails. What is the
conditional probability that the outcome of the toss was heads given that
a white ball was selected?
Solution: Let W be the event that a white ball is drawn, and let H be the
event that the coin comes up heads. The desired probability P pH |W q
may be calculated as follows
P pH X W q
P pH |W q P pW q
P pWP|HpWqPq pH q
P pW |H qP pH q
P pW |H qP pH q P pW |H c qP pH c q
21
21
92
5 1 22
67
92 11 2

Suppose
n that F1 , F2 , ..., Fn are mutually exclusive events such that
i 1 i
F S. In other words, exactly one of the events F1 , F2 , ..., Fn will
occur.nBy writing
E i1 EFi and using the fact that the events EFi , i 1, 2, ..., n, are
mutually exclusive, we obtain that
¸
n
P pE q P pEFi q

i 1
¸n
P pE |Fi qP pFi q

i 1
Suppose now E has occurred and we are interested in determining which

one of the Fj also occurred. We have
P pFj |E q P EFj
P pE q
P pE |Fi qP pFi q
°n
i1 P pE |Fi qP pFi q
This is called Bayes’ Formula

Homework 3
1. A laboratory blood test is 95% effective in detecting certain disease when

it is, in fact, present. However, the test also yields a ”false positive” result
for 1% of the healthy persons tested. (That is if a healthy person is
tested, then, with probability 0.01, the test result will imply he has the
disease.) If 0.5% of the population actually has the disease, what is the
probability a person has the disease given that his test result is positive.
2. If a person is lying, the probability that this is correctly detected by the
polygraph is 0.88, whereas if the person is telling the truth, this is
correctly detected with probability 0.86. Suppose we consider a question
for which 99% of all subjects tell the truth. Our polygraph machine says
a subject is lying on this question. What is the probability that the
polygraph is incorrect?

CHAPTER 2 2.1: Random Variables
Random Variables
Random variables are real-valued functions defined on the sample space

(May be discrete, indicator, or continuous).

CHAPTER 2 2.2: Discrete Random Variables
Discrete Random Variables
A discrete random variable take on a countable set of numbers.

We define for a discrete random variable X, the probability mass function
ppaq given by
ppaq P tX au
The probability mass function ppaq is positive for at most a countable
number of values of a. That is, if X assume one of the values x1 , x2 , ...,
then ppxi q ¡ 0, i 1, 2, .... Otherwise ppxq 0, for other values of x.
Since X must take on one of the values xi , we have
8̧
ppxi q 1.

i 1

The Bernoulli Random Variable
Suppose that a trial, or an experiment, whose outcome can be classified

as either a ”success” or as a ”failure” is performed. If we let X 1 if the
outcome is a success and 0 if it is a failure, then the probability mass
function of X is given by
pp0q P tX 0u 1 p
pp1q P tX 1u p
where p, 0 ¤ p ¤ 1, is the probability that the trial is a success.

The Binomial Random Variable
Suppose that n independent trials, each of which results in a ”success”

with probability p and in a ”failure” with probability 1 p, are to be
performed. If X represents the number of successes that occur in the n
trials, then X is said to be a binomial random variable with parameters
pn, pq.
The probability mass function of a binomial random variable having
pn, pq is given by
parameters
ppiq ni pi p1 pqni , i 0, 1, 2, ..., n

where ni pnn!iq!i! .

The Geometric Random Variable
Suppose that independent trials, each having probability p of being a

success, are performed until a success occurs.
If we let X be the number of trials required until the first success, then X
is said to be a geometric random variable with parameter p.
Its probability mass function is given by
ppnq P tX nu pp1 pqn1 , n 1, 2, ....

The Poisson Random Variable
A random variable X, taking on one of the values 0, 1, 2, . . . , is said to

be a Poisson random variable with parameter λ, if for some λ ¡ 0,
λ i
ppiq P tX iu e i! λ , i 0, 1, 2, ....
An important property of the Poisson random variable is that it may be
used to approximate a binomial random variable when the binomial
parameter n is large and p is small.

Homework 4
1. Suppose that an airplane engine will fail, when in flight, with probability
1 p independently from engine to engine; suppose that the airplane will
make a successful flight if at least 50 percent of its engines remain
operative. For what values of p is a four-engine plane preferable to a
two-engine plane?
2. Consider an experiment that consists of counting the number of
α-particles given off in a one-second interval by one gram of radioactive
material. If we know from past experience that, on the average, 3.2 such
α-particles are given off, what is a good approximation to the probability
that no more than two α-particles will appear?

CHAPTER 2 2.3: Continuous Random Variables
Continuous Random Variables
X is a continuous random variable if there exists a nonnegative function

f pxq, defined for all real x P p8, 8q, having the property that for any
set B of real numbers
»
P tX P Bu f pxqdx.
B
The function f pxq is called the probability density function of the random
variable X.

Uniform Random Variable

A random variable is said to be uniformly distributed over the interval (0,
1) if its probability density function is given by
"
f p xq
1, 0 x 1
0, otherwise
In general, we say that Xis a uniform random variable on the interval
pα, β q if its pdf is given by
" 1
f pxq
β α α x β
0 otherwise
Example 1.4: Calculate the cumulative distribution function of a random
variable uniformly distributed over pα, β q.
$
& 0, a¤α
F pxq
a α

β α, α a β
α¥β
%
1,

Exponential Random Variables
A continuous r.v whose pdf is given for some λ ¡ 0, by

"
λeλx , if x ¥ 0
f p xq
0, if x 0
is said to be an exponential r.v with parameter λ.

Gamma Random Variables
A continuous random variable whose density is given by

#
λeλx λx
p qα1 , if x ¥ 0
f pxq p q
Γ α
0, if x 0
for some λ ¡ 0, α ¡ 0 is said to be a gamma r.v with parameters α, λ.

The quantity Γpαq is called the gamma function is defined by
»8
Γp α q ex xα1 dx.
0

Normal Random Variable
We say that X is a normal random variable with parameters µ and σ 2 if

the density of X is given by
pxµq2
f pxq ?1 e 2σ 2 , 8 x 8. (1)
2πσ
The density function is a bell-shaped curve that is symmetric around µ.
If X is normally distributed with parameters µ and σ 2 then Y αX β
is normally distributed with parameters αµ β and α2 σ 2 .

Proof: Suppose that α ¡ 0 and note that FY p.q , the cdf of the r.v Y , is
given by
FY paq P tY ¤ au
P tαX β ¤ au
" *
aβ
P X¤
α

aβ
FX
α
» paβ q{α
pxµq2
?1 e 2σ 2 dx
»
8 2πσ
" *

a
? 1 exp pv pαµ β qq2
dv
8 2πασ 2α2 σ 2
where the last equality is³ obtained by the change of variables v αx β
However, since FY paq 8 fY pv qdv, it follows that the pdf fY p.q is given
a
by " *
fY pv q ?
1
exp
p v pαµ β qq2
, 8 v 8.
2πασ 2α2 σ 2
Hence, Y is normally distributed with parameters αµ β and α2 σ 2 .
CHAPTER 2 2.4: Expectation of a Random Variable
2.4.1: Discrete Case
The expected value of X is defined by

¸
E rX s xppxq
p q¡0
x:p x
Expectation of a Bernoulli Random Variable

Example: Calculate E[X] when X is a Bernoulli r.v with parameter p.
Solution: Since P(0)=1-p, p(1)=p, we have
E rX s 0p1 pq 1ppq p.
Thus, the expected number o successes in a single trial is just the
probability that the trial will be a success.

Expectation of Binomial Random Variable
¸
n
E rX s iP piq

i 0

¸n
p p1 pqni
n i
i

i 0
i
¸n
pi p1 pqni
in!
i1
p n i q!i!

¸
n
pn 1q! pi1 p1 pqni
np
i1
pn iq!pi 1q!
1 n 1
n¸
np pk p1 pqn1k

k 0
k
nprp p1 pqsn1
np,
where the second from the last equality follows by letting k i 1.
Expectation of Geometric Random Variable
8̧
E rX s npp1 pqn1

n 1
8̧
p nq n1 pwhere q 1 pq
n 1
8̧ d
p pqn q
n1
dq

8̧
p
d
qn
dq

n 1
p
d
dq 1 q
q
1
p1 q q2
1
p
.
Expectation of a Poisson Random Variable
8̧ iλi eλ
E rX s
i0
i!
8̧ λi eλ

pi 1q!
i 1
8̧ λi1
λeλ
i 1
pi 1q!
8̧ λk
λeλ
k 0
k!
λeλ eλ
λ
°8
where we use the identity
λk
k 0 k! eλ .
2.4.2: The Continuous Case
For a continuous r.v X »8

E rX s xf pxqdx
8
Expectation of a Uniform Random Variable
»β
E rX s x
α β α
dx
β 2 α2
2pβ αq
β α
2
.

Expectation of an Exponential Random Variable
»8
E rX s xλeλx dx
o
integrating by parts (dv λeλx , u x) yields

»8
xeλx 8

E rX s 0
eλx dx
o
8
eλx
0
λ

0
1
λ
.

Expectation of a Normal Random Variable
»8
pxµq2
E rX s ?1 xe 2σ 2 dx
2πσ 8
Writing x as px µq µ yields
»8 »8
pxµq2 pxµq2
E rX s ?1 px µqe 2σ 2 dx µ?
1
e 2σ 2 dx
2πσ 8 2πσ 8
Letting y x µ leads to
»8 »8
E rX s ?1 ye 2σ2 dx f pxqdx
y2
µ
2πσ 8 8
where f pxq is the ³normal density. By symmetry, the first integral must be 0,
8
and so E rX s µ 8 f pxqdx µ.
Expectation of a Function of a Random Variable

Proposition 2.1 (a) If X is a discrete random variable with probability
mass function ppxq, then for any real-valued function g,
¸
E rg pX qs g pxqppxq
p q¡0
x:p x
(b) If X is a continuous random variable with probability density function

f pxq, then for any real-valued function g,
»8
E rg pX qs g pxqf pxqdx
8
Corollary 2.2 If a and b are constants, then E raX bs aE rX s b
Proof: For discrete case,
¸
E raX bs pax bqppxq
x:p xp q¡0
¸ ¸
a xppxq b ppxq
p q¡0
x:p x p q¡0
x:p x
aE rX s b
Expectation of a Function of a Random Variable
For continuous case,

»8
E raX bs pax bqf pxqdx
8
»8 »8
a xf pxqdx b f pxqdx
8 8
aE rX s b
E rX s is referred as the mean or the first moment of X. The quantity

E rX n s is called the nth moment of X.
By proposition 2.1, we note that
# °
E rX n
s ³ 8x:ppxnq¡0
pq
xn p x , if X is discrete
8 x f pxqdx, if X is continuous
V arpX q E rpX E rX sq2 s.

Variance of the Normal Random Variable

Recall that E rX s µ. We have
V arpX q E rpX » µq2 s
8 p q
?1 px µq2 e
x µ 2
2σ 2 dx
2πσ 8
Substituting y px uq{σ yields
»8
?σ y 2 e
2
V arpX q y2
2 dy
2π 8
Integrating by parts pu y, dv ye y2
2 dy q
8 »8

?σ e
2
V arpX q ye 2
y2 y2
2 dy
2π
8 8
»8
?σ2 y2
e 2 dy
2π 8
σ2
Homework 5
1. Show that V arpX q E rpX µq2 s can be expressed as

V arpX q E rX 2 s µ2 or V arpX q E rX 2 s pE rX sq2 .
2. Find V arpX q
i. Bernoulli random variable.
ii. binomial random variable.
iii. Poisson random variable.
iv. Geometric random variable.
v. uniform random variable.
vi exponential random variable.

CHAPTER 2 2.5: Jointly Distributed Random Variables
Joint Distribution Functions
Given two discrete random variables X and Y , the joint probability mass
function of X and Y is given by
ppx, y q P tX x, Y y u.
of X may be obtained from ppx, y q by
The pmf °
pX pxq y:ppx,yq¡0 ppx, y q.
Similarly,°
pY py q x:ppx,yq¡0 ppx, y q.
We say X and Y are jointly continuous if there exists a function f px, y q,
defined for all real x and y, having the property that for all sets A and B
of real numbers ³ ³
P tX P A, y P B u B A f px, y qdxdy
The marginal
³ 8 pdfs of X and Y given³ 8 by
fX pxq 8 f px, y qdy and fY py q 8 f px, y qdx.

A variation of proposition 2.1 states that if X and Y are random

variables and g is a function of two variables, then
¸¸
E rg pX, Y qs g px, y qppx, y q in the discrete case
y x
»8 »8
g px, y qf px, y qdxdy in the continuous case
8 8
For example, if g pX, Y q X Y, then in the continuous case ,
»8 »8
E rX Ys px y qf px, y qdxdy
8 8
»8 »8 »8 »8
xf px, y qdxdy yf px, y qdxdy
8 8 8 8
E rX s E rY s
The same result holds in the discrete case.
For constants a and b we have E raX bY s aE rX s bE rY s
If X1 , X2 , ..., Xn are n random variables, then for the n constants
a1 , a2 , ..., an
E ra1 X1 a2 X2 ... an Xn s a1 E rX1 s a2 E rX2s ... an E rXn s
Independent Random Variables

The rvs X and Y are said to be independent if, for all a, b,
P tX ¤ a, Y ¤ bu P tX ¤ auP tY ¤ bu
That is F pa, bq FX paqFY pbq
When X and Y are discrete, the condition of independence reduces to
ppx, y q pX pxqpY py q, while if X and Y are jointly continuous,
independence reduces to f px, y q fX pxqfY py q
Proposition 2.3: If X and Y are independent, then for any functions h
and g E rg pX qhpY qs E rg pX qsE rhpY qs
Proof: Suppose that X and Y are joint. Then
»8 »8
E rg pX qhpY qs g pxqhpy qf px, y qdxdy
8 8
»8 »8
g pxqhpy qfX pxqfY py qdxdy
8 8
»8 »8
g pxqfX pxqdx hpy qfY py qdy
8 8
E rg pX qsE rhpY qs
Covariance and Variance of Sums of Random Variables
The covariance- of any two random variables X and Y , denoted by

Cov pX, Y q, is defined by
Cov pX, Y q E rpX E rX sqpY E rY sqs

E rXY Y E rX s XE rY s E rX sE rY ss
E rXY s E rY sE rX s E rX sE rY s E rX sE rY s
E rXY s E rX sE rY s
Note that if X and Y are independent, then by proposition 2.3 it follows

that Cov pX, Y q 0

Example: The joint density function of X, Y is
1 py { q,
f px, y q e x y
,0 x, y 8
y
(a) Verify that the preceding is a joint density function.

(b) Find Cov pX, Y q.
Solution: To show that f px, y q is a joint density function we need to show
it
³ 8is ³nonnegative, which is immediate, and then that
8 f px, yqdydx 1. We prove the latter as follows:
8 8
»8 »8 »8»8
1 py x{yq
f px, y qdydx e dydx
8 8 0
»8
0 y
»8
ey
1 x{y
e dydx
0 0 y
»8
ey dy
0
1

To obtain Cov px, y q, note that the marginal density function of Y is

»8
fY py q ey
1 x{y
e dx ey
0 y
Thus, Y is an exponential random variable with parameter 1. Therefore,

E rY s 1
We compute- E rX s and E rXY s as follows:
»8 »8
E rX s xf px, y qdydx
8 8»
»8 8x
e y ex{y dydx
0 0 y
³8
Now, 0 xy ex{y dx is the expected value of an exponential random
variable ³with parameter p1{y q and thus equal to y. Consequently,
8
E rX s 0 yey dy 1

Also
»8 »8
E rXY s xyf px, y qdydx
8 8 »
»8 8x
yey ex{y dydx
0 0 y
»8
y 2 ey dy
0
Integration by parts (dv ey dy, u y 2 ) gives

»8
E rXY s y 2 ey dy
0
»8
y2 ey|8 0 2yey dy
0
2E rY s 2
Consequently, Cov pX, Y q E rXY s E rX sE rY s 2 p1qp1q 1.

Properties of Covariance
For any random variables X, Y, Z and constant c

1. Cov pX, X q V arpX q
2. Cov pX, Y q Cov pY, X q
3. Cov pcX, Y q c Cov pX, Y q
4. Cov pX, Y Z q Cov pX, Y q Cov pX, Z q
The fourth property generalizes to give the following result:

¸
n ¸
m ¸
n ¸
m
Cov Xi , Yj Cov pXi , Yj q
i 1
j 1
i 1j 1

We can obtain a useful expression for the variance of the sum of random
variables as follows:

¸
n ¸
n ¸
n
V ar Xi Cov Xi , Xj
i 1
i 1
i j
¸
n ¸
n
Cov pXi , Xj q

i 1j 1
¸n ¸
n ¸
Cov pXi , Xi q Cov pXi , Xj q
i 1
i 1j i
¸n ¸
n ¸
V arpXi q 2 Cov pXi , Xj q
i 1
i 1j i
If Xi , i 1, 2, ..., n are independent random variables, we get

¸
n ¸
n
V ar Xi V arpXi q

i 1
i 1

Definition 2.1: If X1 , X2 , ..., Xn are independent

°n and identically
distributed, then the random variable X̄ i1 Xi {n is called the
sampled mean.
Proposition 2.4: Suppose that X1 , X2 , ..., Xn are independent and
identically distributed with expected value µ and variance σ 2 . Then
(a) E rX̄ s µ
(b) V arpX̄ q σ 2 {n
(c) Cov pX̄, Xi X̄ q 0, i 1, 2, ..., n.
°n
Proof: (a) E rX̄ s n i1 E rXi s µ
1
2 °n °n
(b) V arpX̄ q n1 V ar p i1 Xi q 1 2
n V ar pXi q
i 1
σ2
n
(c)
Cov pX̄, Xi X̄ q Cov pX̄, Xi q Cov pX̄, X̄ q

¸
1
Cov Xi Xj , Xi V arpX̄ q
n j i

¸
1
Cov pXi , Xi q
1
Cov Xi , Xj V arpX̄ q
n n
j i
σ2 2
n
σn 0
Joint Probability Distribution of Functions of Random

Variables
Let X1 and X2 be jointly continuous random variables with joint

probability density function f px1 , x2 q.
It is sometimes necessary to obtain the joint distribution of the random
variables Y1 and Y2 that arise as functions of X1 and X2 .
Specifically, suppose that Y1 g1 pX1 , X2 q and Y2 g2 pX1 , X2 q for some
functions g1 and g2 .
Assume that the equations g1 and g2 satisfy the following conditions:
1. The equations y1 g1 px1 , x2 q and y2 g2 px1 , x2 q can be uniquely solved
for x1 and x2 in terms of y1 and y2 with solutions given by, say,
x1 h1 py1 , y2 q, x2 h2 py1 , y2 q.
2. The functions g1 and g2 have continuous partial derivatives at all points
px1 , x2 q and are such that the following 2 2 determinant
Bg1 B g1
J p x1 , x 2 q
BBxg1 BBxg2 Bg1 Bg2 BBxg Bg2 0 at all points px1 , x2 q.
Bx1 Bx2 B x1
1

B x1 B x2
2 2 2


Under these two conditions it can be shown that the random variables Y1
and Y2 are jointly continuous with joint density function given by
fY1 ,Y2 py1 , y2 q fX1 ,X2 px1 , x2 q|J px1 , x2 q|1 (2)
where x1 h1 py1 , y2 q, x2 h2 py1 , y2 q.

Example: If X and Y are independent gamma random variables with
parameters pα, λq and pβ, λq, respectively, compute the joint density of
U X Y and V X {pX Y q.

Solution: The joint density of X and Y is given by
λeλx pλxqα1 λeλy pλy qβ 1

fX,Y px, y q Γpαq Γpβ q
eλpx yq xα1 y β 1
α β
λ
ΓpαqΓpβ q
Now, if g1 px, y q x y, g2 px, y q x{px y q, then

Bg1 Bg1 1, Bg2 y 2 , Bg2 x 2 , and so
Bx By Bx px yq By px y q

J px1 , x2 q x1y
1 1
y
px yq2 x
px y q2
Finally, because the equations u x y, v x{px y q have as their
solutions x uv, y up1 v q, we that
fU,V pu, v q fX,Y ruv, up1 v qsu

λeλu pλuqα β 1 v α1 p1 v qβ 1 Γpα βq
Γpα β q ΓpαqΓpβ q

Hence, X Y and X pX Y q are independent, with X Y having a

gamma distribution with parameters pα β, λq and pX {pX Y qq having
density function
Γpα β q α1
fV pv q
ΓpαqΓpβ q
v p1 vqβ1 , 0 v 1. (3)
This called the beta density with parameters pα, β q.

STAT422 - Slides

Uploaded by

Copyright:

Available Formats

STAT422 - Slides

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

STAT422 - Slides

Uploaded by

Copyright:

Available Formats

STAT 422: Applied Probability Lecture Notes

Botswana International University of Science and Technology1

F. Chipepa February 7, 2024 1 / 51

F. Chipepa February 7, 2024 2 / 51

Sample Space and Event

Sample Space (S) - is the set of all possible outcomes

F. Chipepa February 7, 2024 3 / 51

F. Chipepa February 7, 2024 4 / 51

Probabilities Defined on Events

F. Chipepa February 7, 2024 5 / 51

Given 3 events E, F, and G

F. Chipepa February 7, 2024 6 / 51

A drug has the following information:

F. Chipepa February 7, 2024 7 / 51

Often we want to calculate the probability that an event E occurs given

F. Chipepa February 7, 2024 8 / 51

F. Chipepa February 7, 2024 9 / 51

Two events are said to be independent if

F. Chipepa February 7, 2024 10 / 51

- Let E and F be events.

The result state that the probability of event E is a weighted average of

F. Chipepa February 7, 2024 11 / 51

F. Chipepa February 7, 2024 12 / 51

Suppose now E has occurred and we are interested in determining which

This is called Bayes’ Formula

1. A laboratory blood test is 95% effective in detecting certain disease when

F. Chipepa February 7, 2024 14 / 51

Random variables are real-valued functions defined on the sample space

F. Chipepa February 7, 2024 15 / 51

Discrete Random Variables

A discrete random variable take on a countable set of numbers.

F. Chipepa February 7, 2024 16 / 51

The Bernoulli Random Variable

Suppose that a trial, or an experiment, whose outcome can be classified

F. Chipepa February 7, 2024 17 / 51

The Binomial Random Variable

Suppose that n independent trials, each of which results in a ”success”

F. Chipepa February 7, 2024 18 / 51

The Geometric Random Variable

Suppose that independent trials, each having probability p of being a

F. Chipepa February 7, 2024 19 / 51

The Poisson Random Variable

A random variable X, taking on one of the values 0, 1, 2, . . . , is said to

F. Chipepa February 7, 2024 20 / 51

F. Chipepa February 7, 2024 21 / 51

Continuous Random Variables

X is a continuous random variable if there exists a nonnegative function

F. Chipepa February 7, 2024 22 / 51

Uniform Random Variable

F. Chipepa February 7, 2024 23 / 51

Exponential Random Variables

A continuous r.v whose pdf is given for some λ ¡ 0, by

is said to be an exponential r.v with parameter λ.

F. Chipepa February 7, 2024 24 / 51

Gamma Random Variables

A continuous random variable whose density is given by

for some λ ¡ 0, α ¡ 0 is said to be a gamma r.v with parameters α, λ.

F. Chipepa February 7, 2024 25 / 51

Normal Random Variable

integrating by parts (dv λeλx , u x) yields

V arpX q E rpX E rX sq2 s.

1. Show that V arpX q E rpX µq2 s can be expressed as

Cov pX, Y q E rpX E rX sqpY E rY sqs

Integration by parts (dv ey dy, u y 2 ) gives

If Xi , i 1, 2, ..., n are independent random variables, we get

where x1 h1 py1 , y2 q, x2 h2 py1 , y2 q.

Now, if g1 px, y q x y, g2 px, y q x{px y q, then

fU,V pu, v q fX,Y ruv, up1 v qsu