Chapter 2

Probability Calculations

4.1 Random Experiments and Events

A random experiment is an experiment whose outcome is uncertain. It is an
experiment whose outcome cannot be completely predicted. In other words, an
experiment which could give different results if it is repeated in the same
conditions.

Let ξ, b e a random experiment:

Definition 2.1.1 (Elementary Event).
A possible result of the experience ξ is called " issue ", or “elementary event ", noted ω.
Definition 2.1.2 (Fundamental set or universe). The set of all possible resultsis called
fundamental set or universal set, noted Ω :
Ω = {ω 1 , 𝜔! , . . . , 𝜔" }
Definition 4.1.3 (Event) An Event A is a subset of Ω, noted A ,B , C, it is therefore a
set of elementary events.

Example 1. The experiment ξ consists of rolling a balanced die.

Ω = { 1 ,2 ,3 ,4 ,5 ,6 } .
Here, the elementary events are { 1 } , {2}, { 3 } , {4} , {5} , {6} . Moreover, if we are
interested in the event E “obtain an even number”, then E = {2, 4, 6}.

Example 3. In a factory, three machines M 1 , M 2 , M 3 , produce daily 300 stems of the

same size. One of them makes brass rods(stems), the two others make steel rods. At the
end of the day, the pieces are mixed.
a) A piece is drawn at random from the daily production.
• Question: How many possible outcomes for this experiment?
• Answer: This experiment is a random with 300 possible outcomes. Let Ω1 be the
fundamental set (universe) associated with this experience. card(𝛺# ) = 300.

1
 b) We draw a part at random, and we are interested in the following question:
“this part is made by which machine?” ".
• Question: How many possible outcomes for this experiment?

• Answer: This random experiment has three possible outcomes:

𝛺! = {𝑀# , 𝑀! , 𝑀$ }.

c) Let us now look, with this same drawing, at the material of the drawn rod.
• Question: give the fundamental set associated with this experience?
• Answer: This random experiment has two possible outcomes
𝛺$ = {brass, steel}.

Remark: Fundamental spaces (or universes) can be of three types:

1. A finite number of possible outcomes

2. An infinite but countable set of possible outcomes (we can list them as in the
case of integers)
We are talking about a discrete universe and we will define discrete
probability laws for this type of universe.
3. An infinite and uncountable set of possible outcomes (an interval of R for
example). We then speak about continuous universe and we will define
continuous probability laws for this type of universe.
Example 4: The experiment consists of measuring the uptime of a computer. Here
Ω=[0, +∞[ and the event A: "the computer operates at least for a duration t0"
corresponds to the subset A = {t | t0 ≤ t < +∞}.

4.1.1 Special Events

• The impossible event, noted ∅, is the event whose occurrence is impossible

whatever the outcome of the test.

• The certain event, noted Ω, is always accomplished it is made up of all

eventualities.

• The opposite or complementary event of an event A, denoted A,

is the event that is realized if and only if A is not. It is therefore constituted
of the elementary events, ω, which aren’t in A.

𝐴̅ = {𝑤 ∈ 𝛺, 𝑤 ∉ 𝐴}

2
4.2 Operations on events
• Union: The set of all the elements which belong to A or B or to both is called
union of A and B, denoted A ∪ B:
ω ∈ A ∪ B ⟺ (ω ∈ A or ω∈B)

• Intersection: The set of all elements that belong to both A and B is called the
intersection of A and B, denoted by A ∩ B:
ω ∈ A ∩ B ⟺ (ω ∈ A and ω ∈B)

• Two events A and B are incompatible or disjoint if they cannot be carried out
simultaneously. We then have: A ∩ B = ∅

• Inclusion: An event A leads to an event B if the realization of A implies that

of B. We say that event A is included in event B , And we note A ⊂ B.

• Difference: The difference of two events A and B is noted A\B, it

contains the elements which belong to A but not to B.
ω ∈A\B ⟺ (ω ∈ A and ω ∉ B)
A\B expresses the fact that A is realized and B is not realized.

• Properties: (Morgan relations). We have the following two properties:

1- 7777777
𝐴 ∪ 𝐵 = 𝐴̅ ∩ 𝐵7
2- 7777777
𝐴 ∩ 𝐵 = 𝐴̅ ∪ 𝐵7

3
4.3 Complete events System

A family (A 1 , A 2, . . . , A n ) of no empty events of Ω form a complete system of

events if and only if it constitutes a partition of Ω :

1. ∀i , ∀j, Ai ∩ Aj =∅

2. 𝛺 = ⋃"%&# 𝐴%

4.4 Probability
Σ
4.4.1 Definition:

The probability P is a real-valued function whose domain is the set of all events (subsets of
Ω) and the range is the interval [0, 1], i.e. P: P(Ω) → [0, 1] satisfying the following axioms.

(i) For any event A , P (A) ≥ 0.

(ii) P(Ω) = 1
(iii) If A and B are two incompatible events, then P(A ∪ B) = P(A) + P(B).
More generally if ( A 1 , A 2 , . .., A n ) is a sequence of two by two incompatible
events then :
𝒏
𝒏
𝑷(= 𝑨𝒊 ) = @ 𝑷( 𝑨𝒊 )
𝒊&𝟏
𝒊&𝟏

4
 Proprieties:

1. P(∅) = 0

A ) = 1 −P( A ) .
2. P ( 𝐴

3. If A ⊂ B, then P ( A ) ≤ P ( B )

A ) = P( A ) −P( A ∩B )
4. P ( A\B ) = P ( A ∩𝐵

5. P ( A ∪ B ) = P ( A ) + P ( B ) − P ( A ∩ B )

6. P (A ∪B∪C)=P (A) + P ( B )+P(C)−P (A∩B)−P (A∩C)−P (B ∩ C)+P (A∩B∩C)

Demonstration:

Property 1 : As ∅ ∩ ∅ = ∅ , we have :

P (∅ ∪ ∅ ) = P ( ∅ ) + P ( ∅ )=2P(∅) =0 ⇒ P(∅) =0

Property 2 :
We know that:
A ∪ A = Ω and A∩A = ∅
Thus we have P (Ω) = P (A ∪ A) = P (A) + P (A)
But, by definition, P (Ω) = 1. Then we deduce:
A ) = 1- P(A).
P (𝑨

Property 3: As A ⊂ B, and we can write the set B as: B=(B∩𝐴A) ∪ 𝐴, with,

(B ∩A) ∩ A = ∅ (compatibility).
So,
P( B ) = P( B ∩A ) + P( A )

And since:

P ( B ∩ A ) ≥ 0 then we deduce that P ( A ) ≤ P( B)

Property 4 : we have A = (A ∩ B) ∪ (A\B) with, (A ∩ B) ∩ (A\B) = ∅(incompatibility)

So: P (A) = P (A ∩ B) + P (A\B)

Then: P (A\B) = P (A) − P (A ∩ B)

5
 Property 5 :
A ∪ B = (A\B ) ∪ ( A ∩ B ) ∪ (B\A )
Moreover, ( A \ B ) , ( A ∩ B ) , And ( B \ A ) are two by two incompatible, SO :
P ( A ∪ B ) = P ( A \ B ) + P ( A ∩ B ) + P ( B\ A )
Or: P (A\B) = P (A) − P (A ∩ B)
P (B\A) = P (B) − P (A ∩ B)
SO : P ( A ∪ B ) = P ( A ) + P ( B ) − P ( A ∩ B)

Property 6 : It is a generalization of property 5: put E=B C then expand:

P(A∪B∪C)=P(A∪E)=………

4.4.2 Equi-probability
Let ω be an element of Ω. If all elementary events {ω} have the same
probability p (equiprobability), then:

𝟏
p=P({ ω })= 𝑵 , 𝑤𝑖𝑡ℎ 𝑁 = |𝛺|.

Therefore, each time we assume equiprobability, we have for every event A:

# ./0123 56 6785/37192 5/:;502<
P(A)=P(⋃+∈- 𝑃({𝑤 })) = ∑+∈- 𝑃({𝑤 }) = ∑+∈- . = =5:79 "/0123 56 5/:;50<

𝒄𝒂𝒓𝒅𝑨
P(A) = 𝑪𝒂𝒓𝒅 𝜴

4.4.3 Frequency
The frequentist probability f of an event A refers to the definition of probability
as the quotient between the number k of favorable outcomes and the number N of
total outcomes when N tends to infinity. Mathematically this is expressed by:

𝑘
𝑃(𝐴) = 𝑙𝑖𝑚
.→FG 𝑁

6
Example: if 60% of men are right-handed, we would say that a man taken at
random from a population has a probability of 0.6 to be right-handed or has 60%
of luck to be right-handed.

4.5 Conditional Probability

If we roll a balanced six-sided die, then the probability of getting the number 4 is of
1/6. If, on the other hand, we have a priori information that the result is even then
the probability of getting the number 4 will be 1/3.
Definition 4.5.1 Let A and B be two events of a random experiment with
fundamental set Ω and P (B)≠0. The conditional probability of an event A given B, is
defined by the following expression:
𝑃(𝐴 ∩ 𝐵)
𝑃(𝐴|𝐵) =
𝑃(𝐵)

• 𝑃(A|B) is the probability that A will occur given that B has occurred

Proposal 4.5.2

Let B an event such P ( B ) > 0 . The application: 𝐴 → 𝑃(𝐴|𝐵) is a probability on 𝛺, it is noted

by P ( . ⁄𝐵) and is called conditional probability knowing B (or given B). We therefore
have the classic properties of a probability. Especially, we have:
A|B) = 1 − P (A|B)
P(𝐴

Be careful !! P( 𝐴| 7𝐵
777) ≠ 1 − P (A| B)

• From the definition of the conditional probability, we deduce that:

P (A ∩ B) = P (A|B) × P (B) = P (B|A) × P (A). (Conditioning)

• For three events A, B and C such as: 𝑃(𝐴) ≠ 0 and 𝑃(𝐴 ∩ 𝐵) ≠ 0, we have:
P (A∩B∩C) = P (A) × P (B|A) × P (C|A∩ B) .

7
 4.5.2 Total Probability Formula
Let { A 1 , A 2, . . . , A n } be a complete events system of Ω

The total probability of an event B defined on the space Ω can be expressed in terms of

conditional probabilities as follows:

𝑃(B) = ∑𝒏𝒊&𝟏 𝑷(𝑩|𝑨𝒊 )𝑷(𝑨𝒊 )

Demonstration: since 𝛺 = (⋃"%&# 𝐴% ) , 𝑡ℎ𝑒𝑛 B=B∩ 𝛺 = 𝐵 ∩ (⋃"%&# 𝐴% ) = (⋃"%&#(𝐵 ∩ 𝐴% )

So, P(B)=P( ⋃"%&#(𝐵 ∩ 𝐴% ))=∑"%&# 𝑃(𝐵 ∩ 𝐴% ) = ∑"%&# 𝑃(𝐵|𝐴% )𝑃(𝐴% )

4.5.3 Bayes Formula

Let { A 1 , A 2, . . . , A n } be a complete events system of Ω .

The Bayes Formula is expressed by : For all event B such as P(B)≠ 0, we have :

𝑷(𝑨𝒊 )𝑷(𝑩|𝑨𝒊 ) 𝑷(𝑨𝒊 )𝑷(𝑩|𝑨𝒊 )

P(𝑨𝒊 |𝑩) = 𝑷(𝑩)
= 𝑨𝒊 ∑𝒏
𝒊#𝟏 𝑷(𝑩|𝑨𝒊 )𝑷(𝑨𝒊 )

Remark: The Bayes rule expresses a conditional probability in terms of other conditional
probabilities.

Example 5.

Two machines 𝑀# 𝑎𝑛𝑑 𝑀! manufacture mechanical parts. The machine 𝑀# produces a third
of the total production. The percentage of defective parts is 5% for 𝑀# and 6% 𝑓𝑜𝑟 𝑀! . Let’s
note by A, B and C, the following events:

A: "the part is manufactured by 𝑀# "

B: "the part is manufactured by 𝑀! "
D: "the part is defective”

8
1. We randomly draw a part in the total production. Calculate the following
probabilities:

A ), 𝑃(𝐵
𝑃(𝐴), 𝑃(𝐵), 𝑃(𝐴 777), P(D|A), P(D|B).

Answer:
# !
A ) =1- 𝑃(𝐴)= ! ; 𝑃(𝐵
P(A) = $ ; P(B) = $ ; 𝑃(𝐴 777)=1- 𝑃(𝐵)= # ; P(D|A ) = N ; P(D|B)= P
$ $ #OO #OO

2. We take a part from the total production. What is the probability, 𝑝# , that it comes
from 𝑀# and it is defective?

Answer:
𝟏 𝟓 𝟏
𝒑𝟏 =P(A∩ 𝑫) = 𝑷(𝑨)𝑷(𝑫|𝑨) = 𝟑 × 𝟏𝟎𝟎 = 𝟔𝟎 =0.0166

3. We take a part from the total production. What is the probability 𝑝! that it is defective?

Answer:
A part being manufactured either by 𝑀# or by 𝑀! , we therefore use
the total probability formula:
𝟏 𝟓 𝟐 𝟔 #V
𝒑𝟐 = P (D) = P ( A ) ×P ( D|A ) + P ( B ) ×P ( D|B ) =𝟑 × 𝟏𝟎𝟎 + 𝟑 × 𝟏𝟎𝟎 = $OO= 0.0566

4. Calculate the probability 𝑝$ for a defective part to have been manufactured by M1?

Answer: By Bayes Formula, we have:

𝑃(𝐴)𝑃(𝐷|𝐴 ) 1 300 5
𝒑𝟑 = 𝑃(𝐴|𝐷) = = × = = 𝟎. 𝟐𝟗𝟒
𝑃(𝐷) 60 17 17

4.5.4 Independent Events

Let A and B two events such as P(A)≠ 𝟎 𝒂𝒏𝒅 𝑷(𝑩) ≠ 𝟎
Definition 1: A and B are said to be independent if and only if
P(A ∩ B) = P(A) × P(B).

9
Remark 1: Independence is then the fact that the achievement (realization) of one
does not depend on the achievement of the other. So, if A and B are two independent
events then: P(B|A) = P(B) and P(A|B) = P(A)

Remark 2: Do not confuse independent events (the occurrence of one does not
modify the probability of the other) and incompatible events (the occurrence of one
prevents the occurrence of the other).

Properties: A and B are independent ⟺ 𝐴7 and B are independent

A are independent
⟺A and 𝐵
⟺ 𝐴7 and 𝐵
A are independent

Definition 2: Mutually independent events:

Three events A, B, and C are mutually independent, if they are independent two by two and

P( A ∩B ∩C ) = P( A ) ×P( B ) ×P( C ) .

Example 5(following). We draw part manufactured by M1 and another manufactured

by M 2 .

1. What is the probability 𝑝W that both parts are defective?

Answer: Let C and E be the following events:

C: "The part is defective and is manufactured by 𝑀# "

E: "The part is defective and is manufactured by 𝑀! ."

The realization of C doesn’t influence that of E. Therefore, C and E are two independent
events. So,
N P
𝒑𝟒 = P(C∩E) = P(C)×P(E) = P(D|A) P(D|B) = #OO × #OO = 𝟎. 𝟎𝟎𝟑2.

2.What is the probability 𝑝N that both parts are non-defective?

Answer: 𝑝N = P(𝐶̅ ∩𝐸7 ).

Since C and E are independent then the events 𝐶̅ and 𝐸7 are also
independent. So, we have:

𝒑𝟓 =P(𝐶̅ )×P(𝐸7 )=(1−P(C))×(1−P(E))=0.95×0.94 = 0.89

10