Probability Discrete random variables

Must
Outcomes
be familiar with
Descriptive statistics

Probability
Probability Discrete random variables

Outcomes
Know the definitions

Identify the link between a sample space and a random variable

Construct pmf /CDF table and chart from word problems

Use pmf /CDF tables /charts to calculate probabilities

Calculate random variable parameters

Solve problems containing linear transformations of random variables

Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times…

X:Ω→ℝ Multiplication rule for probability

S Prob
In words:
× ½ (HHH) ⅛
• X assigns a
real number × ½ (HHT) ⅛
to the ½
outcomes in ½ × ½ (HTH) ⅛
the sample
½
½ (THH) ⅛
space
• Random
× ½ (HTT) ⅛
× ½
variable ⇒ ½ ½ (THT) ⅛
Population
× ½
• Discrete: (TTH) ⅛
countable
½
number of
½ (TTT) ⅛
outcomes
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times…

Cannot perform calculations
X:Ω→ℝ with S (no numbers)

S Prob
In words:
(HHH) ⅛
• X assigns a
real number (HHT) ⅛
to the
outcomes in (HTH) ⅛
the sample (THH) ⅛
space
• Random (HTT) ⅛
variable ⇒
(THT) ⅛
Population
• Discrete: (TTH) ⅛
countable
number of (TTT) ⅛
outcomes
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails
Random variable from sample space
X:Ω→ℝ
X=x P(X = x) S Prob
In words:
0 ⅛ (HHH) ⅛
• X assigns a
real number 1 ⅜ (HHT) ⅛
to the 2 ⅜
outcomes in (HTH) ⅛
the sample 3 ⅛ (THH) ⅛
space
• Random (HTT) ⅛
variable ⇒
(THT) ⅛
Population
• Discrete: (TTH) ⅛
countable
number of (TTT) ⅛
outcomes
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails
Random variable Now we have numbers for calculations
X:Ω→ℝ
X=x P(X = x)
In words:
0 ⅛
• X assigns a
real number 1 ⅜
to the 2 ⅜
outcomes in
the sample 3 ⅛
space
• Random
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails
Random variable
X:Ω→ℝ Like histogram
X=x P(X = x) P(X ≤ x) Probability mass function
In words: 3/8
0 ⅛ ⅛
• X assigns a
real number 1 ⅜ ½
to the 2 ⅜ ⅞ 1/4

Probability
outcomes in
the sample 3 ⅛ 1
space
1/8
• Random Probability
variable ⇒ Mass
Population Function, 0
• Discrete: i.e., pmf 0 1 2 3
countable Number of tails
number of Like relative
outcomes frequencies
for samples
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails
Random variable
X:Ω→ℝ
X=x P(X = x) P(X ≤ x)
In words:
0 ⅛ ⅛
• X assigns a
real number 1 ⅜ ½
to the 2 ⅜ ⅞
outcomes in
the sample 3 ⅛ 1
space
• Random Probability Cumulative
variable ⇒ Mass Distribution
Population Function, Function,
• Discrete: i.e., pmf i.e., CDF
countable
number of Like relative Like ogive for
outcomes frequencies samples
for samples
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails
Random variable
X:Ω→ℝ
X=x P(X = x) P(X ≤ x)
In words:
0 ⅛ ⅛ 𝑃 𝑋 3
• X assigns a
real number 1 ⅜ ½
to the 2 ⅜ ⅞
𝑃 𝑋 2

outcomes in
the sample 3 ⅛ 1 𝑃 𝑋 2
𝑃 𝑋 3

space
𝑃 𝑋 1
• Random Probability Cumulative 𝑃 𝑋 1
variable ⇒ Mass Distribution
Population Function, Function, 𝑃 𝑋 0 𝑃 𝑋 0

• Discrete: i.e., pmf i.e., CDF

countable
number of Like relative Like ogive for
outcomes frequencies samples
for samples
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails
Random variable
X:Ω→ℝ
X=x 𝑝 𝑥 𝐹 𝑥
In words:
0 ⅛ ⅛ 𝑝 3
• X assigns a
real number 1 ⅜ ½
to the 2 ⅜ ⅞
𝑝 2

outcomes in
the sample 3 ⅛ 1 𝐹 2
𝐹 3

space
𝑝 1
• Random Probability Cumulative 𝐹 1
variable ⇒ Mass Distribution
Population Function, Function,
𝑝 0 𝐹 0

• Discrete: i.e., pmf i.e., CDF

countable
number of Like relative Like ogive for
outcomes frequencies samples
for samples
Probability Discrete random variables

Definition In general
These functions can be presented as a table or as an equation
X:Ω→ℝ
𝑋 𝑥 𝑃 𝑋 𝑥 𝑝 𝑥 𝑃 𝑋 𝑥 𝐹 𝑥
Properties of 𝑃 𝑥
In words:
𝑥 𝑝 𝑥 𝑝 𝑥
• X assigns a • 0 𝑃 𝑥 1 when
real number 𝑥 𝑝 𝑥 𝑝 𝑥 𝑝 𝑥
to the 𝑥∈𝑋
𝑥 𝑝 𝑥 𝑝 𝑥 𝑝 𝑥 𝑝 𝑥
outcomes in • 1 ∑∀ 𝑃 𝑥
the sample ⫶ ⫶ ⫶
space Properties of 𝐹 𝑥
• Random Probability Cumulative • 𝑃 𝑋 𝑠𝑚𝑎𝑙𝑙𝑒𝑠𝑡 𝑥 0
variable ⇒ Mass Distribution
Population Function, Function, • 𝑃 𝑋 𝑙𝑎𝑟𝑔𝑒𝑠𝑡 𝑥 1
• Discrete: i.e., pmf i.e., CDF • 𝑃 𝑋 𝑥 𝑃 𝑋
countable
number of Like relative Like ogive for 𝑥 𝑃 𝑋 𝑥 1
outcomes frequencies samples
for samples
Probability Discrete random variables

Definition Distribution parameters

X:Ω→ℝ
𝑋 𝑥 𝑃 𝑋 𝑥
Mean = 𝜇 𝐸𝑋 ∑∀ 𝑥 𝑝 𝑥
In words:
𝑥 𝑝 𝑥
• X assigns a
real number 𝑥 𝑝 𝑥 Population parameter / random variable parameter
to the 𝑥 𝑝 𝑥
outcomes in Variance = 𝑉 𝑋 𝜎 𝐸 𝑋 𝜇
the sample ⫶ ⫶
space
Variance = 𝑉 𝑋 𝜎 𝐸𝑋 𝜇
• Random Variance = 𝑉 𝑋 𝜎 ∑∀ 𝑥 𝑝 𝑥 𝜇
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails

X:Ω→ℝ

Mean = 𝜇 𝐸𝑋 ∑∀ 𝑥 𝑝 𝑥
In words:
• X assigns a
real number
to the
outcomes in X=x P(X = x) 𝑥𝑝 𝑥
the sample
space 0 ⅛ 0× ⁄ = 0
• Random 1 ⅜ 1× ⁄ = ⁄
variable ⇒
Population 2 ⅜ 2× ⁄ = ⁄
• Discrete: 3 ⅛ 3× ⁄ = ⁄
countable
number of 𝐸𝑋 = ⁄ = 1.5 tails are expected when flipping a fair
outcomes coin three times
Probability Discrete random variables

Definition E.g.: Consider flipping a fair coin three times… and let X = number of tails

X:Ω→ℝ
Variance = 𝑉 𝑋 𝜎 𝐸 𝑋 𝜇
In words: Variance = 𝑉 𝑋 𝜎 𝐸𝑋 𝜇
• X assigns a
real number
Variance = 𝑉 𝑋 𝜎 ∑∀ 𝑥 𝑝 𝑥 𝜇
to the
outcomes in X=x P(X = x) 𝑥 𝑝 𝑥 𝑥𝑝 𝑥 𝜎
the sample
space 0 ⅛ 0 × ⁄ = 0 0× ⁄ = 0 𝜎
• Random 1 ⅜ 1 × ⁄ = ⁄ 1× ⁄ = ⁄ 𝜎
variable ⇒
Population 2 ⅜ 2 × ⁄ = ⁄ 2× ⁄ = ⁄
𝜎
• Discrete: 3 ⅛ 3 × ⁄ = ⁄ 3× ⁄ = ⁄
countable
number of 𝐸𝑋 = ⁄ 𝐸𝑋 = ⁄
outcomes
Probability Discrete random variables – Linear transformations

Definition E.g.: Consider the following 5 measurements: 𝑋: 76, 72, 65, 70, 82

𝑋: 76, 72, 65, 70, 82 𝜇 73, 𝜎𝑋 5.7

X:Ω→ℝ
𝑋 3 𝑌1: 79, 75, 68, 73, 85 𝜇 76 𝜇 3, 𝜎𝑌1 5.7 𝜎𝑋
In words: 2𝑋 𝑌2: 152, 144, 130, 140, 164 𝜇 146 2𝜇 , 𝜎 11.5 2𝜎𝑋
• X assigns a
real number 2𝑋 3 𝑌3: 155, 147, 133, 143, 167 𝜇 149 2𝜇 3, 𝜎 11.5 2𝜎𝑋
to the
outcomes in
the sample
space
• Random
variable ⇒
Population
60 70 80 90 100 110 120 130 140 150 160 170
• Discrete:
countable
number of
outcomes
Probability Discrete random variables – Linear transformations

Definition E.g.: Consider the following 5 measurements: 𝑋: 76, 72, 65, 70, 82

𝑋: 76, 72, 65, 70, 82 𝑥̅ 73, 𝑠𝑋 6.4

X:Ω→ℝ
𝑋 3 𝑌1: 79, 75, 68, 73, 85 𝑦 76 𝑥̅ 3, 𝑠𝑌1 6.4 𝑠𝑋
In words: 2𝑋 𝑌2: 152, 144, 130, 140, 164 𝑦 146 2𝑥̅ , 𝑠 12.8 2𝑠𝑋
• X assigns a
real number 2𝑋 3 𝑌3: 155, 147, 133, 143, 167 𝑦 149 2𝑥̅ 3, 𝑠 12.8 2𝑠𝑋
to the
outcomes in
the sample
space
• Random
variable ⇒
Population
60 70 80 90 100 110 120 130 140 150 160 170
• Discrete:
countable
number of
outcomes
Probability Discrete random variables – Linear transformations

Definition In general:

Let 𝑋 be a random variable with 𝐸 𝑋 𝜇 and variance 𝑉 𝑋 𝜎 𝐸𝑋 𝐸𝑋

X:Ω→ℝ
Let 𝑎 and 𝑏 be two real constants
In words: Then 𝐸 𝑎𝑋 𝑏 𝑎𝐸 𝑋 𝑏 𝑎𝜇 𝑏 and 𝑉 𝑎𝑋 𝑏 𝑎2𝑉 𝑋 𝑎2𝜎
• X assigns a
real number Also, if 𝑌 is another random variable with 𝐸 𝑌 𝜇 and variance 𝑉 𝑌 𝜎
to the
outcomes in Distributive
the sample property of E
space Then 𝐸𝑋 𝑌 𝜇 𝜇 𝜇 𝐸𝑋 𝐸𝑌 (always)
• Random
and 𝜎 𝜎 𝜎
variable ⇒
(only when 𝑋 and 𝑌 are independent)
Population 𝜎 𝜎 𝜎
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition Example: Consider 𝑃 𝑋 𝑥 𝑐𝑥, where 𝑥 1,2,3,4.

a. What value of c will make the function a p.m.f.?

X:Ω→ℝ

In words: b. Determine 𝑃 𝑋 2 .
• X assigns a
real number
to the
outcomes in c. Determine 𝑃 1 𝑋 3 .
the sample
space
• Random d. Calculate 𝐸 𝑋 , 𝑉 𝑋 , and the standard deviation of 𝑋.
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition Example: Consider 𝑃 𝑋 𝑥 𝑐𝑥 ⁄ 𝑥, where 𝑥 1,2,3,4.

a. What value of c will make the function a p.m.f.?

X:Ω→ℝ

In words: b. Determine 𝑃 𝑋 2 .
• X assigns a
real number
to the
outcomes in c. Determine 𝑃 1 𝑋 3 .
the sample
space
• Random d. Calculate 𝐸 𝑋 , 𝑉 𝑋 , and the standard deviation of 𝑋.
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition Example: Consider 𝑃 𝑋 𝑥 𝑐𝑥 ⁄ 𝑥, where 𝑥 1,2,3,4.

a. What value of c will make the function a p.m.f.?

X:Ω→ℝ

In words: b. Determine 𝑃 𝑋 2 .
• X assigns a
real number
to the
outcomes in c. Determine 𝑃 1 𝑋 3 .
the sample
space
• Random d. Calculate 𝐸 𝑋 , 𝑉 𝑋 , and the standard deviation of 𝑋.
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition Example: Consider 𝑃 𝑋 𝑥 𝑐𝑥 ⁄ 𝑥, where 𝑥 1,2,3,4.

e. Graph the p.m.f. f. Graph the CDF.

X:Ω→ℝ

In words:
• X assigns a
real number
to the
outcomes in
the sample
space
• Random
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition Example: Consider 𝑃 𝑋 𝑥 𝑐𝑥 ⁄ 𝑥, where 𝑥 1,2,3,4.

e. Graph the p.m.f. f. Graph the CDF.

X:Ω→ℝ

In words:
• X assigns a
real number
to the
outcomes in
the sample
space
• Random
variable ⇒
Population
• Discrete:
countable
number of
outcomes
Probability Discrete random variables

Definition Example: Consider 𝑃 𝑋 𝑥 𝑐𝑥 ⁄ 𝑥, where 𝑥 1,2,3,4.

X:Ω→ℝ

In words: Also consider the discrete random variable Y with 𝐸 𝑌 2 and 𝑉 𝑌 3

• X assigns a
real number If X and Y are independent, then
to the
outcomes in 𝜇
the sample
space 𝜎
• Random and
variable ⇒
Population 𝜇
• Discrete:
countable 𝜎
number of
outcomes
Probability Discrete random variables

Example: Let the random variable X be the number of days that a certain patient needs to be in the hospital.

Suppose X has the p.m.f. 𝑝 𝑥 , where 𝑥 1,2,3,4.

a) What is the probability a patient stays at least 2 days?

b) What is the expected number of days a patient stays in hospital?

c) If the patient will receive R2000 from an insurance

company for each of the first two days in the hospital

and R1000 for each day after the first two days, what

is the expected payment for the hospitalization?

Probability Discrete random variables

Example: Let the random variable X be the number of days that a certain patient needs to be in the hospital.

Suppose X has the p.m.f. 𝑝 𝑥 , where 𝑥 1,2,3,4.

a) What is the probability a patient stays at least 2 days?

b) What is the expected number of days a patient stays in hospital?

c) If the patient will receive R2000 from an insurance

company for each of the first two days in the hospital

and R1000 for each day after the first two days, what

is the expected payment for the hospitalization?

Probability Special discrete random variables – Binomial

Must
Outcomes
be familiar with
Counting techniques: Combinations

Discrete random variables

Probability Special discrete random variables – Binomial

Outcomes
Know the definitions

Know the properties of a binomial experiment

Solve binomial problems using the probability tables and software

Calculate the parameters for a binomial distribution

Solve binomial word problems

Probability Special discrete random variables – Binomial

Examples…

 A selected item may be good or

defective.
 A student may pass or fail an
examination.
 A person selected may be a smoker
or not.
 The price of gold may be up or down.
Probability Special discrete random variables – Binomial

Consider an experiment where…

the same trial is repeated n independent times,

 A selected item may be good or
defective. each trial can result in ONLY a success OR a failure,

 A student may pass or fail an 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

examination. 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

 A person selected may be a smoker
𝑋 the number of successes in n trials, then
or not.
𝑋 is a binomial random variable with parameters 𝑛 & 𝑝
 The price of gold may be up or down.
Probability Special discrete random variables – Binomial

Consider an experiment where…

the same trial is repeated n independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 𝑥 𝑝 1 𝑝 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

where 𝑥 0,1,2,3, … , 𝑛 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in n trials, then

𝐸𝑋 𝜇 𝑛𝑝 and
𝑋 is a binomial random variable with parameters 𝑛 & 𝑝
𝑉 𝑋 𝜎 𝑛𝑝 1 𝑝
Probability Special discrete random variables – Binomial

Effect of parameters 𝑛 and 𝑝 𝑃 𝑋 𝑥 𝑝 1 𝑝 , where 𝑥 0,1,2,3, … , 𝑛

𝐸𝑋 𝑛𝑝 2.5 𝐸𝑋 𝑛𝑝 7.5
Small 𝑛 Small 𝑛 𝑀𝑜𝑑𝑒 8
𝑀𝑜𝑑𝑒 2
𝑀𝑜𝑑𝑒 𝑀𝑒𝑎𝑛 Small 𝑝 Large 𝑝 𝑀𝑒𝑎𝑛 𝑀𝑜𝑑𝑒
⇒ 𝑃𝑜𝑠𝑖𝑡𝑖𝑣𝑒𝑙𝑦 ⇒ Binomial ⇒ Binomial ⇒ 𝑁𝑒𝑔𝑎𝑡𝑖𝑣𝑒𝑙𝑦
𝑠𝑘𝑒𝑤𝑒𝑑 positively negatively 𝑠𝑘𝑒𝑤𝑒𝑑
skewed skewed

𝑛 30 is considered large

𝐸𝑋 𝑛𝑝 𝑀𝑒𝑎𝑛 𝑀𝑜𝑑𝑒 𝐸𝑋 𝑛𝑝 50 0.25 12.5

10 0.5 Large 𝑛 𝑀𝑜𝑑𝑒 12
⇒ 𝑆𝑦𝑚𝑚𝑒𝑡𝑟𝑖𝑐𝑎𝑙 Any 𝑛
5 Small 𝑝 𝑀𝑜𝑑𝑒 𝑀𝑒𝑎𝑛
𝑝 0.5 ⇒ Binomial ⇒ 𝐴𝑝𝑝𝑟𝑜𝑎𝑐ℎ𝑖𝑛𝑔
𝑀𝑜𝑑𝑒 5 𝑠𝑦𝑚𝑚𝑒𝑡𝑟𝑦
⇒ Binomial approaches
symmetric symmetry
Probability Special discrete random variables – Binomial
E.g.1: If the number of independent trials is 𝑛 5, the probability of a success remains 𝑝 0.25 for every trial,
and 𝑋 # of successes, then what is 𝑃 𝑋 2 ?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 𝑥 𝑝 1 𝑝 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

Probability Special discrete random variables – Binomial
E.g.1: If the number of independent trials is 𝑛 5, the probability of a success remains 𝑝 0.25 for every trial,
and 𝑋 # of successes, then what is 𝑃 𝑋 2 ?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 2 0.25 1 0.25 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 2 0.2637 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

Probability Special discrete random variables – Binomial
E.g.1: If the number of independent trials is 𝑛 5, the probability of a success remains 𝑝 0.25 for every trial,
and 𝑋 # of successes, then what is 𝑃 𝑋 2 ?

𝑃 𝑋 2 0.25 1 0.25
𝑃 𝑋 2 0.2637
Probability Special discrete random variables – Binomial
E.g.1: If the number of independent trials is 𝑛 5, the probability of a success remains 𝑝 0.25 for every trial,
and 𝑋 # of successes, then what is 𝑃 𝑋 2 ?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 2 0.25 1 0.25 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 2 0.2637 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

𝑋 2
𝑛 5 𝑋 2
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑝 pmf
Probability Special discrete random variables – Binomial
E.g.1: If the number of independent trials is 𝑛 5, the probability of a success remains 𝑝 0.25 for every trial,
and 𝑋 # of successes, then what is 𝑃 𝑋 2 ?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 2 0.25 1 0.25 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 2 0.2637 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝐸𝑋 𝑛𝑝 5 0.25 1.25 𝑋 is a binomial random variable with parameters 𝑛 & 𝑝
𝑉 𝑋 𝑛𝑝𝑞 5 0.25 0.75 0.9375
Probability Special discrete random variables – Binomial
E.g.2: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 𝑥 𝑝 1 𝑝 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

Probability Special discrete random variables – Binomial
E.g.2: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

𝑃 𝑋 15 0.30 1 0.30
𝑃 𝑋 15 0.00004
Probability Special discrete random variables – Binomial
E.g.2: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

𝑃 𝑋 15 0.30 1 0.30
𝑃 𝑋 15 0.00004

𝑋 15
𝑛 20 𝑋 15
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑝 pmf
Probability Special discrete random variables – Binomial
E.g.2: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

𝑃 𝑋 15 0.30 1 0.30
𝑃 𝑋 15 0.00004

𝐸𝑋 𝑛𝑝 20 0.30 6
𝑉 𝑋 𝑛𝑝𝑞 20 0.30 0.70 4.2
Probability Special discrete random variables – Binomial
E.g.3: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 15 𝑃 𝑋 0 𝑃 𝑋 1 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 15 𝑃 𝑋 2 𝑃 𝑋 3 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑃 𝑋 15 𝑃 𝑋 4 𝑃 𝑋 5
𝑋 the number of successes in 𝑛 trials, then
𝑃 𝑋 15 ⋯
𝑋 is a binomial random variable with parameters 𝑛 & 𝑝
𝑃 𝑋 15 𝑃 𝑋 14 𝑃 𝑋 15
𝑃 𝑋 15 1.0000
From Binomial table
Probability Special discrete random variables – Binomial
E.g.3: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

CDF
table
𝑛 𝑥
parameter value
𝑃 𝑋 15 𝑃 𝑋 0 𝑃 𝑋 1 p parameter
𝑃 𝑋 15 𝑃 𝑋 2 𝑃 𝑋 3
𝑃 𝑋 15 𝑃 𝑋 4 𝑃 𝑋 5
𝑃 𝑋 15 ⋯ CDF
values
𝑃 𝑋 15 𝑃 𝑋 14 𝑃 𝑋 15
𝑃 𝑋 15 1.0000
From Binomial table
Probability Special discrete random variables – Binomial
E.g.3: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

𝑃 𝑋 15 𝑃 𝑋 0 𝑃 𝑋 1
𝑃 𝑋 15 𝑃 𝑋 2 𝑃 𝑋 3
𝑃 𝑋 15 𝑃 𝑋 4 𝑃 𝑋 5
𝑃 𝑋 15 ⋯
𝑃 𝑋 15 𝑃 𝑋 14 𝑃 𝑋 15
𝑃 𝑋 15 1.0000
From Binomial table
Probability Special discrete random variables – Binomial
E.g.3: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

𝑃 𝑋 15 1.0000
From Binomial table

𝑋 15
𝑛 20 𝑋 15
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑝 CDF
Probability Special discrete random variables – Binomial
E.g.4: If the number of independent trials is 𝑛 20, the probability of a success remains 𝑝 0.30 for every
trial, and 𝑋 # of successes, then what is 𝑃 𝑋 15 ?

𝑃 𝑋 15 1.0000
From Binomial table

𝑃 𝑋 15 1 𝑃 𝑋 15
𝑃 𝑋 15 1 1.0000
𝑃 𝑋 15 0
Probability Special discrete random variables – Binomial
E.g.5: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of scoring exactly 50% for the test?
Probability Special discrete random variables – Binomial
E.g.5: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 𝑥 𝑝 1 𝑝 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

Probability Special discrete random variables – Binomial
E.g.5: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 5 ⁄ 1 ⁄ 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 5 0.2461 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

Probability Special discrete random variables – Binomial
E.g.5: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

𝑃 𝑋 5 ⁄ 1 ⁄
𝑃 𝑋 5 0.2461
Probability Special discrete random variables – Binomial
E.g.5: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 5 ⁄ 1 ⁄ 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 5 0.2461 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

𝑋 5
𝑋 5
𝑛 10 pmf
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑝
Probability Special discrete random variables – Binomial
E.g.5: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 5 ⁄ 1 ⁄ 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 5 0.2461 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

How many answers do you expect to guess
𝑋 the number of successes in 𝑛 trials, then
right?
𝐸𝑋 𝑛 𝑝 10 5 𝑋 is a binomial random variable with parameters 𝑛 & 𝑝
Probability Special discrete random variables – Binomial
E.g.6: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of failing the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 5 𝑃 𝑋 0 𝑃 𝑋 1 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 5 𝑃 𝑋 2 𝑃 𝑋 3 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑃 𝑋 5 𝑃 𝑋 4
𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

Probability Special discrete random variables – Binomial
E.g.6: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of failing the test?

𝑃 𝑋 5 𝑃 𝑋 0 𝑃 𝑋 1
𝑃 𝑋 5 𝑃 𝑋 2 𝑃 𝑋 3
𝑃 𝑋 5 𝑃 𝑋 4
𝑃 𝑋 4 0.3770

𝑋 4
𝑋 4 𝑛 10 CDF
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑝
Probability Special discrete random variables – Binomial
E.g.7: You guess the answers to a 10 question, multiple choice test, where each question has two options and
only one option is correct. What is the probability of passing the test?

𝑃 𝑋 4 0.3770
𝑃 𝑋 5 1 𝑃 𝑋 4
𝑃 𝑋 5 1 0.3770
𝑃 𝑋 5 0.6230
Probability Special discrete random variables – Binomial
E.g.8: You guess the answers to a 10 question, multiple choice test, where each question have five options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 5 ⁄ 1 ⁄ 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 5 0.0264 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

𝑋 the number of successes in 𝑛 trials, then

𝑋 is a binomial random variable with parameters 𝑛 & 𝑝

𝑋 5
𝑋 5
𝑛 10 pmf
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑝
Probability Special discrete random variables – Binomial
E.g.8: You guess the answers to a 10 question, multiple choice test, where each question have five options and
only one option is correct. What is the probability of scoring exactly 50% for the test?

the same trial is repeated 𝑛 independent times,

each trial can result in ONLY a success OR a failure,

𝑃 𝑋 5 ⁄ 1 ⁄ 𝑝 𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 probability of a success,

𝑃 𝑋 5 0.0264 𝑞 1– 𝑝 𝑃 𝑓𝑎𝑖𝑙𝑢𝑟𝑒 probability of a failure

How many answers do you expect to guess
𝑋 the number of successes in 𝑛 trials, then
right?
𝐸𝑋 𝑛 𝑝 10 2 𝑋 is a binomial random variable with parameters 𝑛 & 𝑝
Probability Special discrete random variables – Poisson

Must
Outcomes
be familiar with
Discrete random variables

Binomial distribution
Probability Special discrete random variables – Poisson

Outcomes
Know the definitions

Know the properties of a Poisson experiment

Solve Poisson problems using the probability tables and software

Calculate the parameters for a Poisson distribution

Solve Poisson word problems

Use the Poisson to solve binomial problems through estimation

Probability Special discrete random variables – Poisson

Examples…

 The number of bad cheques that arrive

at a bank every day.
 The number of vehicle deaths per
month.
 The number of defective
manufactured items during a month.
 The number of bacteria in a given
culture.
Probability Special discrete random variables – Poisson

Consider an experiment where…

counting events,

events occur in fixed time interval /area /volume,

𝑃 𝑋 𝑥 for 𝑥 0,1,2,… time intervals /areas /volumes are non-overlapping,

!
the observations are independent from each other
where λ is the expected number of
where λ events per unit interval 𝑋 # events in intervals /areas /volumes, then
where e is (Euler’s) mathematical
where e constant 𝑋 is a Poisson random variable with parameter λ

𝐸𝑋 𝑉 𝑋 λ
Probability Special discrete random variables – Poisson
average number of events per interval = 10/7 ≈ 1.4 = λ

(10) Independent events

x x x x x xx xx x
0 1 2 3 4 5 6 7 time
in
hours
(7) Disjoint intervals
Fixed size intervals
Probability Special discrete random variables – Poisson
average number of events per interval = 10/7 ≈ 1.4 = λ

Poisson (λ≈1.4)
4
3
3
intervals
Freq of

2
2
1 1
1
0
0 1 2 3
Events per interval
Probability Special discrete random variables – Poisson
average number of events per area = 10/7 ≈ 1.4 = λ

(10) Independent events

x x x
x xx x x
x
x
0 1 2 3 4 5 6 7 area

(7) Disjoint areas

Fixed size areas
Probability Special discrete random variables – Poisson
average number of events per area = 10/7 ≈ 1.4 = λ

Poisson (λ≈1.4)
4
3
3
Freq of

2
areas

2
1 1
1
0
0 1 2 3
Events per area
Probability Special discrete random variables – Poisson

Effect of parameter λ

→ For small λ the pmf is positively skewed

→ As λ increases, pmf approaches symmetry about the mean

Probability Special discrete random variables – Poisson
E.g.1: If X is a Poisson random variable with an average rate of events 𝜆 5, then what is 𝑃 𝑋 2 ?

𝑃 𝑋 𝑥 for 𝑥 0,1,2,…
!
Probability Special discrete random variables – Poisson
E.g.1: If X is a Poisson random variable with an average rate of events 𝜆 5, then what is 𝑃 𝑋 2 ?

𝑃 𝑋 2
!

𝑃 𝑋 2 0.0842

𝑋 2
𝜆 5 𝑋 2
pmf
Probability Special discrete random variables – Poisson
E.g.2: If X is a Poisson random variable with an average rate of events 𝜆 3.5, then what is 𝑃 𝑋 2 ?

𝑃 𝑋 𝑥
!

𝑃 𝑋 2 0.0842
Probability Special discrete random variables – Poisson
E.g.2: If X is a Poisson random variable with an average rate of events 𝜆 3.5, then what is 𝑃 𝑋 2 ?

. .
𝑃 𝑋 2
!

𝑃 𝑋 2 0.1850

𝑋 2
𝜆 3.5 𝑋 2
pmf
Probability Special discrete random variables – Poisson
E.g.3: If X is a Poisson random variable with an average rate of events 𝜆 15, then what is 𝑃 𝑋 10 ?

𝑃 𝑋 𝑥
!

𝑃 𝑋 2 0.1850
Probability Special discrete random variables – Poisson
E.g.3: If X is a Poisson random variable with an average rate of events 𝜆 15, then what is 𝑃 𝑋 10 ?

𝑃 𝑋 10 𝑃 𝑋 0 𝑃 𝑋 1
𝑃 𝑋 10 𝑃 𝑋 2 𝑃 𝑋 3
𝑃 𝑋 10 ⋯ 𝑃 𝑋 10
Probability Special discrete random variables – Poisson
E.g.3: If X is a Poisson random variable with an average rate of events 𝜆 15, then what is 𝑃 𝑋 10 ?

CDF
table

𝑃 𝑋 10 𝑃 𝑋 0 𝑃 𝑋 1
Different
𝑃 𝑋 10 𝑃 𝑋 2 𝑃 𝑋 3 λ
values
𝑃 𝑋 10 ⋯ 𝑃 𝑋 10 Different
𝑥 Cumulative
values probabilities
Probability Special discrete random variables – Poisson
E.g.3: If X is a Poisson random variable with an average rate of events 𝜆 15, then what is 𝑃 𝑋 10 ?

Different
λ
values

Different Cumulative
𝑃 𝑋 10 𝑃 𝑋 0 𝑃 𝑋 1 𝑥 probabilities
𝑃 𝑋 10 𝑃 𝑋 2 𝑃 𝑋 3 values

𝑃 𝑋 10 ⋯ 𝑃 𝑋 10
Probability Special discrete random variables – Poisson
E.g.3: If X is a Poisson random variable with an average rate of events 𝜆 15, then what is 𝑃 𝑋 10 ?

𝑃 𝑋 10 𝑃 𝑋 0 𝑃 𝑋 1
𝑃 𝑋 10 𝑃 𝑋 2 𝑃 𝑋 3
𝑃 𝑋 10 ⋯ 𝑃 𝑋 10
𝑃 𝑋 10 0.118

𝑋 10
𝜆 15 𝑋 10
CDF
Probability Special discrete random variables – Poisson

E.g.1: The average number of clients arriving per hour at a certain bank is 12. Let X = # clients arriving per

hour, then 𝑋~𝑃𝑜𝑖 12 , i.e., 𝑃 𝑋 𝑥 . Calculate

!

a) 𝑃 𝑋 4 counting events,
b) 𝑃 𝑋 11
events occur in fixed time interval /area /volume,
c) 𝑃 𝑋 11
d) 𝑃 𝑋 11 time intervals /areas /volumes are non-overlapping,

the observations are independent from each other

𝑋 # events in intervals /areas /volumes, then

𝑋 is a Poisson random variable with parameter λ

Probability Special discrete random variables – Poisson

E.g.1: The average number of clients arriving per hour at a certain bank is 12. Let X = # clients arriving per

hour, then 𝑋~𝑃𝑜𝑖 12 , i.e., 𝑃 𝑋 𝑥 . Calculate

!

a) 𝑃 𝑋 4

b) 𝑃 𝑋 11

c) 𝑃 𝑋 11

d) 𝑃 𝑋 11
Probability Special discrete random variables – Poisson

E.g.1: The average number of clients arriving per hour at a certain bank is 12. Let X = # clients arriving per

hour, then 𝑋~𝑃𝑜𝑖 12 , i.e., 𝑃 𝑋 𝑥 . Calculate

!

a) 𝑃 𝑋 4

b) 𝑃 𝑋 11

c) 𝑃 𝑋 11

d) 𝑃 𝑋 11
Probability Special discrete random variables – Poisson estimation to Binomial
𝑋~𝑏 𝑛, 𝑝 E.g.: An insurance company has 1,000,000 clients. Each client has a 0.001% chance of
𝐸𝑋 𝑛 𝑝 dying of a certain disease in any given year. The disease is not contagious but genetic.

if For each client that dies of this disease the company has to pay R100,000.

LARGE 𝑛 and
SMALL 𝑝
then a) What is the probability of 2 clients dying in any given year?

𝑋 ⩪ 𝑃𝑜𝑖 𝑛 𝑝
𝐸𝑋 λ
𝑛 𝑝
b) What is the expected payout for any given year?