Probability Theories: 6301Math3

By: Ali Laksaci

Program: Master of Science in Mathematics,

Department of Mathematical Science,
College of Science, King Khalid University

January 20, 2024

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 1King
/ 44 Kh
 Chapter 1: Probability measures and Probability spaces

1 General probability spaces,

2 Fields and sigma-Fields
3 Probability Spaces,
4 Probability on Finite Sample Spaces,
5 Product Probability Spaces

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 2King
/ 44 Kh
 Introduction to chapter 1

In this chapter, we introduce the fundamental notions of the probability

theory. We note that the most of these notions are given in the course of
Measures theory.

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 3King
/ 44 Kh
 General probability spaces

Definition
The set of all possible outcomes of a random experiment is called a
sample space and is denoted by Ω

Example
An experiment consists of tossing one coin. The sample space is

Ω = {T , H}

Example
An experiment consists of tossing two coins. The sample space is

Ω = {HH, TH, HT , TT }

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 4King
/ 44 Kh
 General probability spaces

Definition
Let Ω be a sample space (nonempty set). A collection F of subsets of Ω is
called field or algebra of Ω if:
1 F is nonempty collection
if (Ei )ni=1 ∈ F then ni=1 Ei ∈ F ( F closed under finite union )
S
2

3 if E ∈ F then E c ∈ F ( F closed under complement )

Remark
1 Ω∈F
2 ∅∈F
3 The algebra F is closed under finite intersection

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 5King
/ 44 Kh
 General probability spaces

Example
F = {∅, Ω} is an algebra in Ω, because:
1 F is nonempty?, yes, F contains two elements
2 F closed under finite union ? yes, Ω ∪ ∅ = ∅ ∪ Ω = Ω ∈ F
3 F closed under complement ? yes, Ωc = ∅ ∈ F and ∅c = Ω ∈ F
Therefore, all the conditions are satisfied.

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 6King
/ 44 Kh
 General probability spaces

Definition
Let Ω be a nonempty set. A collection F of subset of Ω is called σ-field
or σ-algebra of Ω if:
1 F is nonempty collection
S∞
2 if (Ei )∞
i=1 ∈ F then i=1 Ei ∈ F ( F closed under countable union)
3 if E ∈ F then E c ∈ F ( F closed under complement )

In what follows, we suppose that F is σ-algebra

Definition

The pair (Ω, F) is called measurable space and the elements of F are
called measurable sets
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 7King
/ 44 Kh
 General probability spaces

Example
Let (Ω, F) a given measurable space, check that
Ω is measurable set
∅ is measurable set
If A and B measurable sets, then, A ∩ B measurable set
If A and B measurable sets, then, A/B measurable set ?
If A and B measurable sets, then, A∆B measurable set ?

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 8King
/ 44 Kh
 General probability spaces

Theorem
Suppose that (Fi )i∈I is a non-empty collection (indexed by I ) of σ-algebras
Fi on Ω. Then the intersection F = ∩i Fi is also σ-algebra on Ω.

Theorem
Let C be a class of subsets of Ω, there exist a unique smallest σ-algebra,
denoted σ(C), contains C.

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science, 9King
/ 44 Kh
 General probability spaces

Definition
The σ-algebra σ(C) is called σ- algebra generated by C.

Remark
σ(C)= The intersection of all σ- algebra containing C

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,10King
/ 44 Kh
 General probability spaces

Example

What is the σ- algebra generated by the class C = {A} ? is

σ(C) = {A, Ω, ∅, Ac }

What is the σ- algebra generated by the class C = {A, B}

σ(C) = {A, B, Ω, ∅, Ac , B c , A ∪ B A ∩ B, (A ∪ B)c (A ∩ B)c , . . .}

if Ω = IR the σ- algebra generated by the intervals is called Borel σ-

algebra in IR and is denoted BIR

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,11King
/ 44 Kh
 General probability spaces

Definition
Let F be a topology in Ω, the σ- algebra generated by the collection of the
open sets in F is called Borel σ- algebra in Ω

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,12King
/ 44 Kh
 General probability spaces

Theorem
If Ω = R, the Borel σ-algebra is generated by any of the following
collections of subsets of R:
(i) C1 = {] − ∞, x], x ∈ R}
(ii) C2 = {]x, y ], x < y ∈ R}
(iii) C3 = {[x, y ], x < y ∈ R}

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,13King
/ 44 Kh
 General probability spaces

Definition
Let (Ω, F) be a measurable space. A measure µ is a function

µ : F 7→ [0, +∞]
A 7→ µ(A)

such that
1 ∀A ∈ F µ(A) ≥ µ(∅) = 0
2 σ-additivity: For all disjoint countable sets
(Ai )∞
i=1 (Ai ∩ Aj = ∅, if i ̸= j)

we have
∞ ∞
!
[ X
µ Ai = Ai .
i=1
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Departmenti=1
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,14King
/ 44 Kh
 General probability spaces

Definition

The triple (Ω, F, µ) is called measure space.

Definition
If µ(Ω) = 1, µ is called probability measure, and the triple (Ω, F, µ) is
called probability space.

In this course, we focus in this last case ( µ(Ω) = 1 ). So, in what

follows the probability measure µ is denoted by P and the
probability space is (Ω, F, P)

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,15King
/ 44 Kh
 General probability spaces

Example
Suppose that Ω is a finite set and consider the trivial σ-algebra

F = P(Ω): the class of all subsets of Ω,

check that
µ(·) = #(·) is probability measure ?
#(·)
P(·) = is probability measure?
#(Ω)

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,16King
/ 44 Kh
 General probability spaces

Theorem

Let (Ω, F, P) be a probability space

1 For all A, B ∈ F

P(A ∪ B) = P(A) + P(B) − P(A ∩ B)

2 If A ⊂ B then

P(A) ≤ P(B) and P(B/A) = P(B) − P(A)

3 For all A, B ∈ F

P(A∆B) = P(A) + P(B) − 2P(A ∩ B)

By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,17King
/ 44 Kh
 General probability spaces

Theorem

Let (Ω, F, P) be a probability space

1 If (An )n a sequence of subsets in F
n n
!
[ X
P An ≤ P(An )( is called Bonferroni inequality’s )
n=1 n=1

2 If (An )n a sequence of subsets in F such that ∀ n, An ⊂ An+1 then

!
[
P An = sup(P(An )) = lim P(An )
n≥1 n→∞
n=1

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,18King
/ 44 Kh
 General probability spaces

Theorem
1 If (An )n a sequence of subsets in F such that ∀ n, An+1 ⊂ An then
!
\
P An = inf (P(An )) = lim P(An )
n≥1 n→∞
n=1

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,19King
/ 44 Kh
 General probability spaces

Definition
A collection M of subsets of Ω is called monotone class if
1 Ω∈M
2 If A and B are in M with B ⊂ A then A/B is also ∈ M
3 If (An )n a sequence of subsets in F such that ∀ n, An ⊂ An+1 then
S
n=1 An ∈ M.

Definition
Let C be a class of subsets of Ω, there exist a unique smallest monotone
class, denoted M(C), contains C and is called monotone class generated by
C.

Remark
M(C)= The intersection of all monotone class containing C
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,20King
/ 44 Kh
 General probability spaces

Definition
Let C be a class of subsets of Ω. If C, is closed under finite intersection, is
called π-system of Ω.

Theorem
Uniqueness of probability measure
Let P1 and P2 be two probability measures on (Ω, F). If, for all subset A
in a π-system C, we have P1 (A) = P2 (A), then P1 = P2 in σ(C) the
σ-algebra generated by C.

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,21King
/ 44 Kh
 General probability spaces

For the Proof we use the following Lemma,

Lemma
Dynkin lemma
Let C be a class of subsets of Ω. If C is π-system, then,

M(C) = σ(C) ( monotone class generated by C = the σ-algebra generated

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,22King
/ 44 Kh
 General probability spaces

Proof of the Theorem

We define M = {A such that P1 (A) = P2 (A)}. It is clear that
C∈M
Now, we show that M is monotone class. Indeed,
The first condition is Ω in M is verified ?, Yes, because
P1 (Ω) = 1 = P2 (Ω). So Ω ∈ M.
Next for the seconde condition, we have to show that
If A and B are in M with B ⊂ A then A/B is also ∈ M? (1)
Indeed, since, A and B are in M then P1 (A) = P2 (A) and
P1 (B) = P2 (B). On the other hand, as B ⊂ A therefore
P1 (A/B) = P1 (A) − P1 (B) = P2 (A) − P2 (B)
it follows that
P1 (A/B) = P2 (A/B)
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,23King
/ 44 Kh
 General probability spaces

Thus, the second condition (1) is also verified.

Finally, we check the last one that is
If (An )n a sequence of subsets in F such that ∀ n, An ⊂ An+1
S
then n=1 An ∈ M.
Indeed, from the item 4 in Theorem (20), we have
!
[
P1 An = sup(P1 (An )) = lim P1 (An )
n≥1 n→∞
n=1

and !
[
P2 An = sup(P2 (An )) = lim P2 (An ).
n≥1 n→∞
n=1
As for all n the subset An is ∈ M, then
∀n P1 (An ) = P2 (An )
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,24King
/ 44 Kh
 General probability spaces

It follows that
lim P1 (An ) = lim P2 (An ).
n→∞ n→∞

We deduce that ! !
[ [
P1 An = P2 An .
n=1 n=1

So, the third condition is also satisfied.

Hence, M is a monotone class contains C . So M ⊃ M(C), but
M(C) = σ(C) implies that , M ⊃ σ(C). Consequently for all A ∈ σ(C) we
have P1 (A) = P2 (A).

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,25King
/ 44 Kh
 General probability spaces

Definition
Let P be a probability measure defined on (IR, BIR ). The distribution
function (f.d.) of P is defined by

F (x) = P (] − ∞, x]) .

Proposition
P1 = P2 if and only if the distribution function F1 of P1 EQUAL the
distribution function F2 of P2

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,26King
/ 44 Kh
 General probability spaces

Proof of the proposition

We have to show that :
(I) P1 = P2 ⇒ F1 = F2

(II) F1 = F2 ⇒ P1 = P2

The first assertion is trivial. Observe that ∀x ∈ IR,

F1 (x) = P1 (] − ∞, x]) = P2 (] − ∞, x]) = F2 (x)
So, F1 equal to F2
Now, we suppose that for all x ∈ IR, F1 (x) = F2 (x). So, we consider
the class
C = (] − ∞, x])x∈IR .
This class is a π-system of IR, in sense that, this class is closed under
finite intersection. Indeed, let x1 ≤ x2
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,27King
/ 44 Kh
 General probability spaces

(] − ∞, x1 ]∩] − ∞, x2 ] =] − ∞, x1 ] ∈ C.
Moreover, let a1 ≤ a2 ≤ a3 . . . an then

(] − ∞, a1 ]∩] − ∞, a2 ] . . . ∩] − ∞, an ] =] − ∞, a1 ] ∈ C.

Now, we apply the Theorem of the uniqueness of probability measure to

deduce that P1 = P2 in the σ-algebra generated by the class C. This
equality is also true for the Borel σ- algebra in IR. Because, the Borel σ-
algebra in IR include in the σ-algebra generated by the class C. Indeed, by
the definition the Borel σ- algebra in IR is generated by the class of the
intervals ([a, b])a,b∈IR . As, all interval [a, b] can be written as

[a, b] =] − ∞, b]/] − ∞, a].

the σ(C) contains al the interval of IR. Finally, σ(C) contains the
σ-algebra generated by the intervals which is Borel σ- algebra in IR.
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,28King
/ 44 Kh
 General probability spaces

Proposition
Let P be a probability measure in (IR, BIR ) and if F its distribution
function. This function has the following properties:
1 F is increasing function
2 F is a right continuous function
3 limx→−∞ F (x) = 0 and limx→+∞ F (x) = 1
4 For all a, b such that a < b we have P(]a, b]) = F (b) − F (a)
5 P(] − ∞, a[) = F (a− ) and P(a) = F (a) − F (a− )
6 F is continuous only and only if P(a) = 0

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,29King
/ 44 Kh
 General probability spaces

Proof of the Proposition

1 For the first claimed result: For all x, y ∈ IR such that x ≤ y , we

have ] − ∞, x] ⊂] − ∞, y ]. Then
P (] − ∞, x]) ≤ P (] − ∞, y ]) .
Thus
F (x) ≤ F (y ).
2 For the second claimed result: It suffices to prove that
lim F (x) = F (a).
x→a+
 
1
To do that, we define a sequence of a subsets An = −∞, a + . It
n
is clear that, for all n ≥ 1 An+1 ⊂ An then from Theorem (20), we
have !
\
P An = inf P(An )) = lim P(An ). (2)
n≥1 n→∞
n=1
Since
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
 Theories:
Department
6301Math3
of Mathematical Science,
 JanuaryCollege
20, 2024
of Science,30King
/ 44 Kh
 General probability spaces

1 For the first part of the third claimed result: Similarly to the
previous case we define another sequences An =] − ∞, −n] ; n ≥ 1. ,
this sequence such that An+1 ⊂ An then from Theorem (20), we have
!
\
P An = inf P(An )) = lim P(An ). (3)
n≥1 n→∞
n=1
T
Observe that n=1 An = ∅ and P(An ) = F (−n) then, equation (3)
implies that
F (−∞) = lim F (−n) = P(∅) = 0.
n→∞
2 For the second part of the third claimed result: Once again we
define another sequences An =] − ∞, +n] ; n ≥ 1. , this sequence
such that An ⊂ An+1 then from Theorem (20), we have
!
[
P An = sup P(An )) = lim P(An ). (4)
n≥1 n→∞
n=1

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,31King
/ 44 Kh
 General probability spaces
S
Now, we use the fact that n=1 An = IR and P(An ) = F (n) then,
equation (3) implies that

F (+∞) = lim F (n) = P(IR) = 1.

n→∞

1 Now, we prove, the fourth claimed result: For this aim, we use
the fact that
]a, b] =] − ∞, b]/] − ∞, a],
therefore, as
] − ∞, a] ⊂] − ∞, b]
then
P (]a, b]) = P (] − ∞, b]/] − ∞, a])
= P (] − ∞, b]) − P (] − ∞, a]) = F (b) − F (a)
In conclusion, we can write P ]a, b] = F (b) − F (a)
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,32King
/ 44 Kh
 General probability spaces

1 For the first part of the fifth claimed result Using, the same
arguments as the previous case and consider a sequences
1
An =] − ∞, a − ] n ≥ 1,
n
this sequence such that An ⊂ An+1 then from Theorem (20), we have
!
[
P An = sup P(An )) = lim P(An ). (5)
n≥1 n→∞
n=1

Now, using the fact that

[ 1
An =] − ∞, a[ and P(An ) = F (a − )
n
n=1

then, equation (5) implies that

1
F (] − ∞, a[) = lim F (a − ) = F (a− ).
n→∞ n
By: Ali Laksaci ( Program: Master of Science inProbability
Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,33King
/ 44 Kh
 General probability spaces

1 For the second part of the fifth claimed result Similarly to the
fourth case, we use the fact that

{a} =] − ∞, a]/] − ∞, a[,

therefore, as
] − ∞, a[⊂] − ∞, a]
then
P ({a}) = P (] − ∞, a]/] − ∞, a[)
= P (] − ∞, a]) − P (] − ∞, a[) = F (a) − F (a− )

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,34King
/ 44 Kh
 General probability spaces

In consequence, we have

P ({a}) = F (a) − F (a− )

1 For the last claimed result, Evidently, F is continuous only and only
if
F (a+ ) = F (a) = F (a− )
which is equivalent to write

P{a} = 0.

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,35King
/ 44 Kh
 General probability spaces

Definition
Let A be a measurable set in probability space (Ω, F, P) such that
P(A) > 0. The conditional probability , given A, is the function denoted
by P(|A) defined by

P(A ∩ B)
∀B ∈ F P(B|A) = .
P(A)

Exercise
Is P(·|A) probability measure?

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,36King
/ 44 Kh
 General probability spaces

Theorem
Multiplicative Theorem
Let A1 , A2 , . . . An be a n measurable sets such that
!
\
P nAi > 0.
i=1

Then
!
\
P nAi = P (An |A1 ∩ A2 ∩ . . . ∩ An−1 )×P (An−1 |A1 ∩ A2 ∩ . . . ∩ An−2 )×
i=1

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,37King
/ 44 Kh
 General probability spaces

Theorem
Total Probability Theorem
Let A1 , A2 , . . . An . . . be a measurable sets such that
[
∀ i ̸= j Ai ∩ Aj ≠ ∅ ∞Ai = Ω and P(Ai ) > 0.
i=1

Then
∞
X
∀B ∈ F P(B) = P(B|Ai )P(Ai ).
i=1

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,38King
/ 44 Kh
 General probability spaces

Theorem
Bayes Formula
Let A1 , A2 , . . . An . . . be a measurable sets such that
[
∀ i ̸= j Ai ∩ Aj ≠ ∅ ∞Ai = Ω and P(Ai ) > 0.
i=1

Then
P(B|Ai )P(Ai )
∀B ∈ F such that P(B) > 0 P(Ai |B) = P∞ .
i=1 P(B|Ai )P(Ai )

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,39King
/ 44 Kh
 General probability spaces

Definition
Suppose that (Ω1 , F1 ) and (Ω2 , F2 ) are measurable spaces. The
σ-algebra, denoted F1 ⊗ F2 on Ω1 × Ω2 generated by

is called the product σ-algebra and the measurable space

(Ω1 × Ω2 , F1 ⊗ F2 ) is called a product space.

Definition
Two measurable sets A and B in a probability space (Ω, F, P) are said to
be independent if
P(A ∩ B) = P(A)P(B).

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,40King
/ 44 Kh
 General probability spaces

Definition
The n measurable sets A1 , A2 , . . . An . . . are said to be completely
independent if
∀ collection Ai1 , Ai2 , . . . Aik
such that P(Ai1 ∩ Ai2 ∩ . . . ∩ Aik ) = P(Ai1 ).P(Ai2 ). . . . P(Aik )).
These measurable sets are said to be pairwise independent if

∀ i ̸= j P(Ai ∩ Aj ) = P(Ai )P(Aj ).

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,41King
/ 44 Kh
 General probability spaces

Exercise
If A and B are independent then
A and B c are independent
Ac and B are independent
Ac and B c are independent

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,42King
/ 44 Kh
 General probability spaces

Definition
Suppose that (Ω1 , F1 , P1 ) and (Ω2 , F2 , P2 ) are probability space. The
σ-algebra, denoted F1 ⊗ F2 on Ω1 × Ω2 generated by the collection

C = {A1 × A2 A1 ∈ F1 and A2 ∈ F2 }

is called the product σ-algebra and the measurable space

(Ω1 × Ω2 , F1 ⊗ F2 ) is called a product space.

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,43King
/ 44 Kh
 General probability spaces

Definition
In the the product space (Ω1 × Ω2 , F1 ⊗ F2 ) we define a probability
measure P = P1 × P2 such that

P(A1 × A2 ) = P1 (A1 )P2 (A2 ).

P is called product probability measure

By: Ali Laksaci ( Program: Master of Science inProbability

Mathematics,
Theories:
Department
6301Math3
of Mathematical Science,
JanuaryCollege
20, 2024
of Science,44King
/ 44 Kh