IIMT 3636 Decision and Risk Analysis

Topic 1 Probability Concepts and

Applications
Dr. Huiyin Ouyang

Faculty of Business and Economics

The University of Hong Kong
Learning Objectives
1 Understand the basic foundations of probability
analysis.

2 Use Bayes’ theorem to establish posterior probabilities.

3 Describe discrete and continuous random variables.

4 Understand the common discrete and continuous

distributions.

5 Understand the normal distribution and use the normal

table.
2
1 Fundamental Concepts of
Probability

• Basic rules of probability

• Determination of Probability
• Mutually Exclusive and Collectively Exhaustive Events
• Conditional Probability and Independent Events

3
Introduction
• Life is uncertain; we are not sure what the future will bring
• Probability is a numerical statement about the likelihood
that an event will occur
– 10% chance of rain tomorrow
– 20% chance the Hang Seng Index will not go down next week
– 30% chance the launch of the new iPhone will be delayed
– 70% chance Donald Trump will win the election again
– > 99.99% chance that the sun will rise tomorrow morning.

• Notation: 𝑃 𝐴 = Probability of event A occurring.

• 0 ≤ 𝑃 𝐴 ≤ 1.
4
Two Basic Rules of Probability
• The probability, P, of any event or state of nature occurring
is greater than or equal to 0 and less than or equal to 1.
That is,
0 ≤ P(event) ≤ 1
A probability of 0 indicates that an event is never expected
to occur. A probability of 1 means that an event is always
expected to occur.
• The sum of the simple probabilities for all possible
outcomes of an activity must equal 1.
• Regardless of how probabilities are determined, they must
adhere to these two rules.
5
Determination of Probability
• Objective Approach
Number of occurrences of the event
P (event) =
Total number of trials or outcomes
– Relative frequency approach (using data)
– Classical or logical method
• Subjective Approach
– Based on the experience and judgment of the person
making the estimate
▪ Expert opinion
▪ Opinion polls (wisdom of the crowd)

6
 Diversey Paint Example

• Calculate the probability of each possible outcome for

future demand. 7
 Diversey Paint Example

Number of occurrences of the event

P (event) =
Total number of trials or outcomes

QUANTITY DEMANDED NUMBER

PROBABILITY
(GALLONS) OF DAYS
0 40 0.20 (= 40÷200)
1 80 0.40 (= 80÷200)
2 50 0.25 (= 50÷200)
3 20 0.10 (= 20÷200)
4 10 0.05 (= 10÷200)
Total 200 1.00 (= 200÷200)
1. Individual probabilities are all between 0 and 1
2. Total of all event probabilities equals 1 8
Determination of Probability
• Objective Approach
– Classical or logical method ▪ Drawing a card from a deck
▪ Tossing a fair coin: 𝑃(head)? of 52 cards: 𝑃(spade)?

1  Number of ways of getting a head

P ( head) =
2  Number of possible outcomes(head or tail)
13  Number of chances of drawing a spade
P ( spade ) =
52  Number of possible outcomes
1
= = 0.25 = 25%
4 9
Mutually Exclusive and Collectively
Exhaustive Events
• Events are said to be mutually exclusive if only one of
the events can occur on any one trial
– Tossing a coin will result in either a head or a tail
– Rolling a die will result in only one of six possible
outcomes

FIGURE 1 Venn Diagram for FIGURE 2 Venn Diagram for Events

Events That Are Mutually Exclusive That Are Not Mutually Exclusive 10
Mutually Exclusive and Collectively
Exhaustive Events
• Events are said to be collectively exhaustive if the list of
outcomes includes every possible outcome
– Head and Tail of tossing a coin

– All six possible outcomes of the roll of a dice

– Odd and Even of the outcomes of the roll of a dice

11
 Drawing a Card Example

• Draw one card from a deck of 52 playing cards

A = event that a King is drawn
B = event that a heart is drawn

P (a King is drawn) = P(A) = 4/52 = 1/13

P (a heart is drawn) = P(B) = 13/52 = 1/4

– These two events are not mutually exclusive since a

King of hearts can be drawn
– These two events are not collectively exhaustive since
there are other cards in the deck besides Kingss and
hearts 12
In-class
Exercises
• Mutually exclusive: Events cannot happen together
• Collectively exhaustive: one of the Events must happen

Mutually Collectively
Two events: A and B
Exclusive Exhaustive
1. Draw a spade and a club
2. Draw a face card and a number card
3. Draw an ace and a 3
4. Draw a club and a non-club
5. Draw a 5 and a diamond
6. Draw a red card and a non-diamond

13
Union and Intersection
• The union of two events is the set of all
outcomes that are contained in either of the A B
two events.
𝑃 Union of 𝐴&𝐵 = 𝑃 𝐴 or 𝐵 = 𝑃 𝐴 ∪ 𝐵

• The intersection of two events is the set of

all outcomes that are common to both
events.
𝑃 Intersection of 𝐴&𝐵 = 𝑃 𝐴 and 𝐵 = A B
𝑃 𝐴 ∩ 𝐵 = 𝑃 𝐴𝐵 ; it is called joint prob.

• Collectively exhaustive: 𝑃(𝐴 ∪ 𝐵) = 1

• Mutually exclusive: 𝑃(𝐴 ∩ 𝐵) = 0
14
 Unions and Intersections Example

• In the previous example

– Intersection of event A and event B
(A and B) = the King of hearts is drawn
P(A ∩ B) = P(King of hearts is drawn) = 1/52

– Union of event A and event B

(A or B) = either a King or a heart is drawn
P(A ∪ B) = P(any King or any heart is drawn) = 16/52

15
Unions and Intersections
• General rule for union of two events,
additive rule

P(A or B) = P(A) + P(B) − P(A and B)

P(A ∪ B) = P(A) + P(B) − P(A ∩ B)

– Union of two events, a King or a heart

P(A or B) = P(A) + P(B) − P(A and B)
= 4/52 + 13/52 − 1/52
= 16/52

16
Conditional Probability
• Conditional probability – probability that event A occurs
given event B has already happened

P ( AB )
P( A | B) =
P (B )
P ( AB ) = P ( A | B )P (B )
– Probability of a King given a heart has been drawn

1
P ( AB )
P( A | B) = = 52 = 1
P (B ) 13 13
52

17
Independent events
• Independent one event has no effect on the other event
P(A | B) = P(A)
P(A and B) = P(A)P(B)
– Denote by 𝐴 ⊥ 𝐵
– For a fair coin tossed twice

A = event that a head is the result of the first toss

B = event that a head is the result of the second toss

P(A) = 0.5 and P(B) = 0.5

P(AB) = P(A)P(B) = 0.5(0.5) = 0.25
18
Summary
• Marginal (simple) probability is just the probability of an event
occurring
P (A)
• Joint probability is the probability of two or more events occurring.
P (AB)

If A and B are independent, P (AB) = P(A) x P(B).

• Conditional probability is the probability of event B given that event
A has occurred
P (B | A) = P (AB)/P(A) for dependent event
= P(B) for independent event
19
Summary
• 0 ≤ 𝑃 𝐴 ≤ 1 for any event 𝐴.

Also, 0 ≤ 𝑃 𝐴|𝐵 ≤ 1, 0 ≤ 𝑃 𝐴𝐵 ≤ 1, 0 ≤ 𝑃 𝐴 ∪ 𝐵 ≤ 1.

• 𝑃 𝐴 ∩ 𝐵 = 0 if 𝐴 and 𝐵 are mutually exclusive.

• 𝑃 𝐴 ∪ 𝐵 = 𝑃 𝐴 + 𝑃 𝐵 , if A and B are mutually exclusive.

• 𝑃 𝐴 ∪ 𝐵 = 1 if 𝐴 and 𝐵 are collectively exhaustive.

• 𝑃 𝐴∪𝐵 +𝑃 𝐴∩𝐵 =𝑃 𝐴 +𝑃 𝐵 .

• 𝑃 𝐴 ∩ 𝐵 = 𝑃 𝐴|𝐵 ∙ 𝑃 𝐵 = 𝑃 𝐵|𝐴 ∙ 𝑃 𝐴 .

• 𝑃 𝐴|𝐵 = 𝑃 𝐴 if 𝐴 and 𝐵 are independent.

20
 In-class Exercises
1. A student taking MS301 at East Haven University will receive one of the
five possible grades for the course: A, B, C, D, or F. The distribution of
grades over the past two years is in the table.
a) What is the probability of a student Number of
Grade
Students
receiving at least C?
A 80
b) Given that a student didn’t fail, what is
the probability of this student receiving C? B 75
C 90
D 30
F 25

2. Suppose 𝐴 and 𝐵 are mutually exclusive and collectively exhaustive. In

addition, 𝑃 𝐴 = 0.4
a) What is 𝑃 𝐵 ?
b) Suppose 𝐶 and 𝐷 are also mutually exclusive and collectively
exhaustive. We have 𝑃 𝐶|𝐴 = 0.2 and 𝑃 𝐷|𝐵 = 0.4. What are
𝑃 𝐶 and 𝑃 𝐷 ?
21
2. Revising Probabilities with
Bayes’ Theorem

22
Bayes’ Theorem
• Bayes’ theorem is used to incorporate additional
information and help create posterior probabilities from
original or prior probabilities
• What is the probability of a person infected by HIV (Human
Immuno-deficiency Virus)?

HK Prevalence Rate
= 0.1%

𝑃 Infected|Test Positive =?
Lab Test 𝑃 Infected|Test Negative =? 23
Bayes’ Theorem
• Test accuracy
– If a person is infected, the test will return positive result
90% of the time
– If a person is NOT infected, the test will return negative
result 95% of the time
• How to revise your probability assessment when you have
new information?
Diagnostic test for the Human Immuno-deficiency Virus (HIV)

Infected Not Infected

Test Positive 90% (conditional)
Test Negative 95% (conditional)
HK Prevalence Rate 0.1% (marginal) 99.9% (marginal) 24
Bayes’ Theorem
• 𝐴 = Infected; 𝐴′ = Not Infected.
– A’ = the complement of the event A (i.e., not A)

• 𝐵 = Test Positive; 𝐵′ = Test Negative.

General Form of Bayes’ Theorem

𝑃 𝐴𝐵 𝑃 𝐵|𝐴 𝑃 𝐴 𝑃 𝐵|𝐴 𝑃 𝐴
• 𝑃 𝐴|𝐵 = = =
𝑃 𝐵 𝑃 𝐵𝐴 +𝑃 𝐵𝐴′ 𝑃 𝐵|𝐴 𝑃 𝐴 +𝑃 𝐵|𝐴′ 𝑃 𝐴′

– Note:𝑃 𝐵 = 𝑃 𝐵𝐴 + 𝑃 𝐵𝐴′ (Law of total probability)

0.9×0.001 =1.77%
• 𝑃 Infected|Test Positive = 0.9×0.001+0.05×0.999
25
Bayes’ Theorem
Tabular Form of Bayes’ Calculations Given That Event B Has
Occurred
STATE OF P(B | STATE PRIOR JOINT POSTERIOR
NATURE OF NATURE) PROBABILITY PROBABILITY PROBABILITY
A P(B | A) ×P(A) =P(B and A) P(B and A)÷P(B) = P(A | B)
A′ P(B | A′) ×P(A′) =P(B and A′) P(B and A′)÷P(B) = P(A′ | B)
Blank Blank Blank P(B) Blank

New Information Event B: Test Positive

STATE OF P(Test Positive PRIOR JOINT POSTERIOR

NATURE | STATE PROBABILITY PROBABILITY PROBABILITY
OF NATURE)

Infected 90% ×0.1% = 0.09% 0.09÷5.085 = 1.77%

Not Infected 5% ×99.9% = 4.995% 4.995÷5.085 = 98.23%
Blank Blank Blank P(Test Positive) Blank
= 5.085%
26
Bayes’ Theorem: Matrix Calculation
• Draw a matrix for possible events related to two types
(dimensions) of information.
Original Information Infected Not Infected
Test Positive 90% (conditional)
Test Negative 95% (conditional)
HK Prevalence Rate 0.1% (marginal) 99.9% (marginal)

• Calculate the joint probabilities

Bayes Calculation Infected Not Infected
Test Positive 0.9*0.001=0.0009 0.05*0.999=0.04995 0.05004
Test Negative 0.1*0.001=0.0001 0.95*0.999=0.94905 0.94906
HK Prevalence Rate 0.1% (marginal) 99.9% (marginal)

• Calculate conditional probabilities 27

3. Random Variables
• Random variables
• Probability distribution functions
• Summary statistics: expected value, variance and
standard deviation

28
Random Variables
• A random variable (RV) assigns a number to every
possible outcome or event in a statistical experiment.
• A discrete RV can assume only a finite or coutable set of
values
• E.g., X = the number of newspapers sold during the
day.
• A continuous RV can assume any one of an infinite set of
values
• E.g., Y = the lifespan of a light bulb.

29
 Random Variables
Examples of Random Variables

EXPERIMENT OUTCOME RANDOM VARIABLES RANGE OF

RANDOM
VARIABLES
Stock 50 Christmas Number of Christmas X = number of Christmas trees sold 0, 1, 2, . . . , 50
trees trees sold
Inspect 600 items Number of acceptable Y = number of acceptable items 0, 1, 2, . . . , 600
items
Send out 5,000 Number of people Z = number of people responding to 0, 1, 2, . . . , 5,000
sales letters responding to the letters the letters

Build an apartment Percent of building R = percent of building completed 0 … R … 100

building completed after 4 after 4 months
months
Test the lifetime of Length of time the bulb S = time the bulb burns 0 … S … 80,000
a lightbulb lasts up to 80,000
(minutes) minutes

30
RV with Qualitative Outcomes
• When the outcome itself is not numerical or quantitative, it
is necessary to define an RV that associates each
outcome with a unique real number. E.g.
– For tossing a coin, X = 1 if head and 0 if tail;
– For consumers’ response to how they like a product,
Y = 1 if poor, 2 if average, and 3 if good;
– For the brand of soda purchased by a consumer, Z = 1
if Pepsi, 2 if Coca-Cola, and 3 if Dr. Pepper.

31
Random Variables
Random Variables for Outcomes That Are Not Numbers

EXPERIMENT OUTCOME RANGE OF RANDOM RANDOM

VARIABLES VARIABLES

Students respond to Strongly agree (SA) 1, 2, 3, 4, 5

a questionnaire 5 if SA
Agree (A) 4 if A

Neutral (N) X = 3 if N
2 if D
Disagree (D) 
1 if SD
Strongly disagree (SD)
One machine is Defective 0 if defective 0, 1
inspected Y =
Not defective 1 if not defective
Consumers Good 3 if good 1, 2, 3
respond to how 
they like a product Average
Z = 2 if average
1 if poor
Poor 

32
Probability Distributions
• For discrete random variables, probability value assigned
to each value
– Quiz with five problems with 1 point for each correct
answer
– Lowest score = 1, highest score = 5
• Follows the three rules:
1. Outcomes are mutually exclusive and collectively
exhaustive
2. Individual probability values between 0 and 1
3. Total probability sums to 1

33
 Probability Distributions

Probability Distribution for Quiz Scores, for a class with 100

students
RANDOM VARIABLE NUMBER PROBABILITY P(X)
(X) SCORE
5 10 0.1 = 10÷100
4 20 0.2 = 20÷100
3 30 0.3 = 30÷100
2 30 0.3 = 30÷100
1 10 0.1 = 10÷100
Total 100 1.0 = 100÷100

• Developed using relative frequency approach

34
Probability Distribution: Discrete RV
• For each possible outcome 𝑋𝑖 , there is a probability value
𝑃 𝑋𝑖 .
• These values must be between 0 and 1: 0 ≤ 𝑃 𝑋𝑖 ≤ 1.
• They must sum up to 1: σ𝑛𝑖=1 𝑃 𝑋𝑖 = 1.
• 𝑃 𝑋𝑖 : probability mass function (pmf)
• 𝐹 𝑥 = 𝑃 𝑋 ≤ 𝑥 = σ𝑋𝑖 ≤𝑥 𝑃 𝑋𝑖
cumulative probability function (cdf)

35
Probability Distribution: Continuous RV
• For each possible interval (𝑥, 𝑥 + Δ), there is a probability
value 𝑃 𝑋 ∈ (𝑥, 𝑥 + Δ) = 𝑓 𝑥 Δ for small Δ
– The probability of each individual value of the random
variable occurring must equal 0
• 𝑓(𝑥): probability density function (pdf)
– Note 𝑓(𝑥) is not a probability (it’s possible 𝑓 𝑥 > 1 )
𝑏
– The probabilities: 𝑃 𝑋 ∈ (𝑎, 𝑏) = ‫𝑓 𝑎׬‬ 𝑥 𝑑𝑥 ∈ 0,1 .
∞
– They must sum up to 1: ‫׬‬−∞ 𝑓 𝑥 𝑑𝑥 = 1
𝑥
• 𝐹 𝑥 = 𝑃 𝑋 ≤ 𝑥 = ‫׬‬−∞ 𝑓 𝑢 𝑑𝑢: cumulative distribution
function (cdf)
36
 Probability Distribution: Continuous RV
• For any continuous Probability Density Function
distribution, the probability
does not change if a
single point is added to
the range of values that is
being considered.

• The following probabilities

are all exactly the same:

𝑃 5.22 < 𝑋 < 5.26 = 𝑃 5.22 < 𝑋 ≤ 5.26 = 𝑃 5.22 ≤ 𝑋 < 5.26
= 𝑃 5.22 ≤ 𝑋 ≤ 5.26 = F 5.26 − F(5.22)

37
Summary Statistics: Expected Value
• Expected value is a measure of the central tendency of
the distribution
• For a discrete random variable:

𝜇 = 𝐸 𝑋 = σ𝑛𝑖=1 𝑋𝑖 ∙ 𝑃 𝑋𝑖
= 𝑋1 ∙ 𝑃 𝑋1 + 𝑋2 ∙ 𝑃 𝑋2 + ⋯ + 𝑋𝑛 ∙ 𝑃 𝑋𝑛
where
𝑋𝑖 = random variable’s possible values
𝑃(𝑋𝑖 ) = probability of each possible value of the random variable
σ𝑛𝑖=1 = summation sign indicating we are adding all 𝑛 possible values
𝐸(𝑋) = expected value or mean of the random variable

38
 Summary Statistics Expected Value Example

• For the quiz scores

n
E ( X ) =  Xi P ( Xi )
i =1

= X1P (X1 ) + X2 P (X2 ) + X3 P (X3 ) + X 4 P (X 4 ) + X5 P (X5 )

= (5)(0.1) + (4)(0.2) + (3)(0.3) + (2)(0.3) + (1)(0.1)
= 2.9
39
Summary Statistics: Variance and
Standard Deviation
• The variance is a measure of the overall spread or
dispersion of a distribution, and it is defined as
𝑛

𝜎 2 = 𝑉 𝑋 = ෍ 𝑋𝑖 − 𝐸 𝑋 2 ∙ 𝑃 𝑋𝑖
𝑖=1
where
𝑋𝑖 = random variable’s possible values
𝐸(𝑋𝑖) = expected value of the random variable
[𝑋𝑖 − 𝐸(𝑋)] = difference between each value of the random
variable and the expected value
𝐸(𝑋) = probability of each possible value of the random
variable

• The standard deviation is the square root of variance.

s = Variance = s 2 40
 Summary Statistics Variance and
Standard Deviation Example
• For quiz scores 𝐸 𝑋 = 2.9
n
Variance =  [Xi − E (X)]2 P (Xi )
i =1

Variance = (5 − 2.9)2(0.1) + (4 − 2.9)2(0.2) + (3 − 2.9)2(0.3)

+ (2 − 2.9)2(0.3) + (1 −2.9)2(0.1)
= (2.1)2(0.1) + (1.1)2(0.2) + (0.1)2(0.3) + (−0.9)2(0.3) +
(−1.9)2(0.1)
= 0.441 + 0.242 + 0.003 + 0.243 + 0.361
= 1.29
Standard deviation 𝜎 = 1.29 = 1.14
41
 Excel Output for the Quiz Example
Using Excel

42
In-class Exercises
• True or False?
– Define X to be a random variable as follows: X = 0 if it rains
tomorrow and X = 1 if it is sunny tomorrow.
– Suppose a random variable X has the following three possible
values: -1, 0, and 1. Then the expectation of X is 0.
– If 𝑓(𝑥) is the pdf of a random variable X, we must have 𝑓 𝑥 ≤ 1.
– If X is a discrete random variable, then the possible values of X
must be integers.
– If X is a discrete random variable that can take integer values from
1 to 5, then 𝑃 2 ≤ 𝑋 ≤ 3 = 𝑃 𝑋 ≤ 3 − 𝑃 𝑋 ≤ 2 .

43
4 Common Distributions

• Binomial Distribution (Discrete)

• Poisson Distribution (Discrete)
• Uniform Distribution (Continuous)
• Exponential Distribution (Continuous)
• Normal Distribution (Continuous)

44
The Binomial Distribution
• The binomial distribution is used to find the probability of a
specific number (𝑟) of successes in 𝑛 trials
𝑛!
𝑃 𝑟; 𝑛, 𝑝 = ∙ 𝑝𝑟 ∙ 1 − 𝑝 𝑛−𝑟
𝑟! 𝑛 − 𝑟 !
n = number of trials
p = the probability of success on any single trial
r = number of successes
The symbol ! means factorial, and n! = n×(n − 1) ×(n − 2)…× 1. For
example: 4! = 4×3×2×1= 24. Also, 1! = 1 and 0! = 0 by definition

• If 𝑋~𝐵 𝑛, 𝑝 , then 𝐸 𝑋 = 𝑛𝑝 and 𝑉𝑎𝑟 𝑋 = 𝑛𝑝 1 − 𝑝

45
 Binomial Distribution Examples
• Toss a fair coin for n times. Let X be the number of heads you get. X ~
Binomial(n, 0.5)
– the probability of getting 4 heads：
5!
– 𝑃(𝑋 = 4) = 0.54 0.55-4
4!(5-4)!
5(4)(3)(2)(1)
= (0.0625)(0.5) = 0.15625
4(3)(2)(1)1!

• A real estate agent serving 30 customers a week. The probability of a

customer buying an apartment is p. Let Y be the number of apartments
sold per week.
– Y~Binomial(30, p)
• You send 100 internship applications. The probability of a successful
application is p. Let Z be the number of offers you receive.
– Z~Binomial(100,p).
46
Binomial Distribution
Binomial Distribution for n = 5, p = 0.50

47
Using Excel
Excel Output for the Binomial Example

48
Poisson Distribution
• A discrete probability distribution
– number of independent arrivals during a unit period of
time.
– Probability density function given by
𝑒 −𝜆 ∙ 𝜆𝑥
𝑃 𝑥; 𝜆 =
𝑥!
Note 𝑃(𝑥; 𝜆) = probability of exactly 𝑥 arrivals or occurrences
𝝀 = average number of arrivals per unit of time
(the mean arrival rate)
e = 2.718, the base of natural logarithms
𝑥 = number of occurrences (0, 1, 2, 3, …)

• If 𝑋~𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝜆 , then 𝐸 𝑋 = 𝜆, 𝑉𝑎𝑟 𝑋 = 𝜆.

49
 Poisson Distribution Examples

• A call center staff on average receives 2 calls per minute.

Let X be the # of calls s/he receives the next minute.
– X ~ Poisson(2).
– If 𝑋 is the # of calls s/he receives during the next 30
minutes, then X∼ Poisson(60). (where arrival rate 60=
2 × 30)
• A part of highway segments encounter 4 accidents per
year on average. Let Y be the # of accidents happening in
this segment of highway in 2021.
– Y ~ Poisson(4).
– If 𝑌 is the # of accidents happening in the first quarter
of 2021, then 𝑌 ∼Poisson(1).

50
Poisson Distribution Examples
Sample Poisson Distributions with λ = 2 and λ = 4

51
Using Excel
Excel Output for the Poisson Distribution

52
Uniform Distribution
• A random variable 𝑋 is uniformly distributed if it takes any
value equally likely from a finite interval 𝑎, 𝑏 .
𝑎+𝑏 𝑏−𝑎 2
• If 𝑋~𝑈 𝑎, 𝑏 , then 𝐸 𝑋 = 2
and 𝑉 𝑋 = 12

Probability density function Cumulative distribution function

53
Exponential Distribution
• The exponential distribution often
describes the time required to f
service a customer or the lifespan
of a product (e.g., a light bulb)

• If 𝑋~𝐸𝑋𝑃 𝜆 , then 𝑓 𝑥 = 𝜆𝑒 −𝜆𝑥

and 𝐹 𝑥 = 1 − 𝑒 −𝜆𝑥 , where 𝜆 =
1/𝐸 𝑋
F
• 𝐸 𝑋 = 1/𝜆 and 𝑉 𝑋 = 1/𝜆2

54
 Exponential Distribution Examples

• A supermarket cashier on average serves 1 customer for 3

minutes. John buys something and goes to this cashier.
Let T be his service time (# of minutes) at the counter.
– 𝑇 ~ Exp(1/3).
– 𝐸 𝑇 = 3 (minutes), 𝑉𝑎𝑟 𝑇 = 9

• Certain type of DNA strands on average remains

unchanged for 0.5 year between two mutates. Suppose
one such DNA strand just mutates. Let X be the time (# of
years) until the next mutation.
– 𝑋 ~ Exp(2)
– 𝐸 𝑋 = 1/2 (years), 𝑉𝑎𝑟 𝑋 = 1/4 55
Exponential and Poisson Distributions
• Exponential and Poisson are related: if the number of
arrival per time period follows Poisson(𝜆0 ), then the time
between arrivals (interarrival time) follows exponential
distribution with rate 𝜆1 = 𝜆0 .

• Example: Suppose customer arrivals at a retail store per

hour follow Poisson with mean 𝜆 = 3. Given that a
customer arrived at 2:30pm, what is the probability that the
next customer will arrive before 3:00pm?

56
Using Excel
Excel Output for the Exponential Distribution

57
5 Normal Distribution

• Normal Distribution
• Standard Normal Distribution
• Using the Standard Normal Table

58
Normal Distribution
• One of the most popular and useful continuous probability
distributions
– E.g., return of a stock portfolio, forecast errors, and test
scores.
– The probability density function
1 𝑥−𝜇 2
−
𝑓 𝑥 = ∙ 𝑒 2𝜎2
𝜎 2𝜋
– Completely specified by the mean, μ, and the standard
deviation, σ
– We often use the notation 𝑁 𝜇, 𝜎 2 .

59
Normal Distribution
Normal Distribution with Different Values for μ

60
Normal Distribution
Normal Distribution with Different Values for σ

61
Normal Distribution
• Symmetrical with the midpoint representing the mean
• Shifting the mean does not change the shape
• Values on the X axis measured in the number of standard
deviations away from the mean
• As standard deviation becomes larger, curve flattens
• As standard deviation becomes smaller, curve becomes
steeper

62
Normal Distribution

63
Standard Normal Distribution

𝑿−𝝁
If 𝑿~𝑵 𝝁, 𝝈𝟐 , then 𝒁 = follows standard normal distribution.
𝝈
64
Using the Standard Normal Table Example

• For μ = 100, σ = 15, 𝑋 ∼ 𝑁 𝜇, 𝜎

• Find the probability that 𝑋 is less than 130

X − 130 − 100
Z= =
 15
30
= = 2 std dev
15

65
Using the Standard Normal Table
Step 1
• Convert the normal distribution into a standard normal
distribution
– Mean of 0 and a standard deviation of 1
– The new standard random variable is Z
X −
Z=

where
X = value of the random variable we want to measure
μ = mean of the distribution
σ = standard deviation of the distribution
Z = number of standard deviations from X to the mean, μ

66
Using the Standard Normal Table
Step 2
• Look up the probability from a table of normal curve areas
• Use Standard Normal Table (Provided in Exam)
• Column on the left is Z value
• Row at the top has second decimal places for Z values

67
Standard Normal Table: Z Table

68
 69

Using the Standard Normal Table Example

AREA UNDER THE NORMAL CURVE

Z 0.00 0.01 0.02 0.03
1.8 0.96407 0.96485 0.96562 0.96638
1.9 0.97128 0.97193 0.97257 0.97320
2.0 0.97725 0.97784 0.97831 0.97882
2.1 0.98214 0.98257 0.98300 0.98341
2.2 0.98610 0.98645 0.98679 0.98713

For Z = 2.00
P(X < 130) = P(Z < 2.00) = 0.97725
P(X > 130) = 1 − P(X ≤ 130) = 1 − P(Z ≤ 2)
= 1 − 0.97725 = 0.02275
Using the Standard Normal Table
• Find a specified percentile
– 𝑋 ∼ 𝑁 𝜇, 𝜎
– Find the 80th percentile of 𝑋: find a value 𝑥 such that
80% of the possible values of 𝑋 are no larger than 𝑥
▪ E.g., 50th percentile of 𝑋 is 𝜇
– Equivalent: Find a number 𝑥 such that 𝑋 will be less
than 𝑥 with probability 0.8, i.e., 𝑃 𝑋 < 𝑥 = 0.8.
– How to find this value 𝑥?

70
Finding the x value
• Step 1: Finding the z value corresponding to 0.8
– If we need 𝑥 for 𝑃 𝑋 < 𝑥 , find 𝑧 corresponding to
𝑃 𝑋<𝑥 .
– 𝑧0.8 = 0.84

• Step 2: Setting 𝑥 to 𝑥𝑧 = 𝜇 + 𝑧 ∙ 𝜎.
𝑋−𝜇
– 𝑃 𝑋 < 𝑥𝑧 = 𝑃 𝑋 < 𝜇 + 𝑧 ∙ 𝜎 = 𝑃 <𝑧 =
𝜎
𝑃 𝑍<𝑧 =𝑃 𝑋<𝑥
– 𝑃 𝑋 < 100 + 𝑧0.8 × 15 = 𝑃 𝑋 < 112.6 = 0.8
71
 Haynes Construction Company

• Builds three- and four-unit apartment buildings

– Total construction time follows a
normal distribution
– For triplexes, μ = 100 days
and σ = 20 days
– Contract calls for
completion in 125 days
– Late completion will
incur a severe
penalty fee
– Probability of completing in 125 days?

72
 Haynes Construction Company

• Compute Z

X − 125 – 100
Z= =
 20
25
= = 1.25
20
𝑃 𝑋 < 125 = 𝑃 𝑍 < 1.25 = 0.8944

The probability is about 0.89

that Haynes will not violate
the contract

73
 Haynes Construction Company
• If finished in 75 days or less, bonus = $5,000
– Probability of bonus?

X − 75 – 100
Z= =
 20
–25
= = –1.25
20
• Because the distribution is
symmetrical: The probability of completing
𝑃 𝑋 < 75 = 𝑃 𝑍 < −1.25 the contract in 75 days or less
= 𝑃 𝑍 > 1.25 is about 11%
= 1 − 𝑃 𝑍 < 1.25
= 1 − 0.8944 = 0.1056 74
 Haynes Construction Company

• Probability of completing between 110 and 125 days?

𝑃(110 < 𝑋 < 125) = 𝑃(𝑋 ≤ 125) − 𝑃(𝑋 < 110)

– 𝑃(𝑋 ≤ 125) = 0.8944

X − 110 – 100
Z= =
 20
10
= = 0.5
20
– 𝑃 𝑋 < 110 = 𝑃 𝑍 < 0.5 = 0.6915
• 𝑃 110 < 𝑋 < 125
= 0.8944 − 0.6915 = 0.2029
75
 Haynes Construction Company **

• Suppose Haynes can hire more workers to reduce the

mean construction time while keeping the standard
deviation constant. Hence, to ensure a penalty is avoided
with probability 0.95, what should be the new mean?

𝑋 ′ −𝜇
• Objective: find 𝜇 such that 𝑃 𝑋′ < 125 = 𝑃 ൬ 20 <
125−𝜇
൰ = 0.95
20

• Step 1: 𝑧0.95 = 1.65

125−𝜇
• Step 2: find 𝜇 such that = 1.64; 𝜇 = 92
20
76
Using Excel
Excel Output for the Normal Distribution Example

77
Appendix:
Standard Normal Table (Part 1)

78
Appendix:
Standard Normal Table (Part 2)

79
Summary
1 Fundamental Concepts

2 Revising Probabilities with Bayes’ Theorem

3 Random Variables and Probability Distributions

4 Common Distributions

‒ Binomial Distribution, Poisson Distribution

‒ Uniform Distribution, Exponential Distribution

5 The Normal Distribution and the Normal Table

80