PROBABILITY THEORY

MTM2592-Gr:2

SECOND WEEK
Assc. Prof. Dr. Ulku (BABUSCU)
YESIL
[email protected]
[email protected]
https://avesis.yildiz.edu.tr/ubabuscu

1
 Probability Concepts
EXPERIMENT:
Statisticians use the word ‘experiment’ to describe any process that
generates a set of data. For example; tossing a coin is an
experiment. It generates ‘Head’ and ‘Tail’. If an experiment always
generates the same outcome under the same conditions, it is called
a deterministic experiment. Otherwise, it is called a random
experiment. In this course we deal with random experiments.
SAMPLE POINT & SAMPLE SPACE:
The set of all possible outcomes of an experiment is called sample
space, and it is usually denoted by the letter S. Each outcome of the
experiment is called a sample point of S. Sample points build n-
dimensional sample space.

S  H, T 
S  x, y  x 1, 2,.., 6; y 1, 2,.., 6 
 (1,1), (1, 2),...., (6, 6)
Sample space Sample point
Sample point
2
DISCRETE AND CONTINUOUS SAMPLE SPACE:
A sample space that contains a finite number of elements or countable
infinite elements is called a discrete sample space. Otherwise, it is called
a continuous sample space.

For example; S 1, 2,3, 4,5, 6 discrete sample space

S  x 2  x  5 continuous sample space

3
4
5
Example: If a random experiment consist of tossing a balanced die, then
the sample space is

S 1, 2,3, 4,5, 6

A1 1,3,5 The event of getting odd numbers

A 2  2, 4, 6 The event of getting even numbers

A 3 1, 2 The event of getting a number smaller than 3

Example: Lifetime of produced bulbs

S  x 0 x  

B1  x x  200 The event of lasting longer than 200 hours

Example: Let a balanced coin be tossed three times. Construct a sample

space and an event that the coin shows head at least two times.
S  HHH, HHT, HTH, HTT, THH, THT, TTH, TTT
A  HHH, HHT, HTH, THH
6
Example: Let S  x 1 x 5 be a sample space and
A  x 1 x 4 , B  x 2  x 5 be two events. Construct the events

A  B  x 1 x 5
A  B  x 1 x 4
A  B  x 4  x 5


A  B  x 1 x 2 or 4  x 5 

7
THE DEFINITIONS OF THE PROBABILITY:
Conventional Definition: Let S be a sample space that contains equally
likely outcomes and A be an event defined on S. The probability of the
event A is
n(A)
P(A) 
n(S)

Ex: What is the probability that a balanced die comes up greater than 4?

n(A) 2
S 1, 2,3, 4,5, 6 A  5, 6 P(A)  
n(S) 6

8
Frequency Definition: If N is the number of trials of an experiment and n
is the number of times the event A happens, then

n
p(A)  lim
N  N

Subjective Definition: The use of intuition, personel beliefs and other

indirect information in arriving at probabilities is referred to as the
subjective definition of probability.
x: The chance of occurrence of an event A
y: The chance of non-occurrence of an event A

p(A) x x
  p(A)  (subjective chance)
1  p(A) y xy

Ex: An expert considers that the chance of Besiktas’s winning the cup is
2:5. According to this expert what is the probability that Besiktas will win
the cup?

P=2/(2+5)=2/7
9
 PROBABILITY AXIOMS:
Definition: A real valued function P, defined on a sample space S is
called a probability function if the following axioms are satisfied.

1) 0 P(A) 1 for all events A

2) P(S)=1

3) If A1, A2,…., An are disjoint events (mutually exclusive), then

P(A1  A 2  A3 ....) P(A1 )  P(A 2 )  P(A3 )  ......

For two disjoint events P(A B) P(A)  P(B)

10
Theorem 1: P( ) 0 where  is an empty set.

Proof: A is any set.

A    and A   A
P(A) P(A  ) P(A)  P( ) P( ) 0

Theorem 2: P(A) 1  P(A) where A is the complement of A.

Proof: S A  A and A  A 

1 P(S) P(A  A) P(A)  P(A)

P(A) 1  P(A)

11
Theorem 3: If A B then P(A) P(B)

Proof: B \ A   A , B \ A   A B
P B \ A   A  P B \ A   P(A) P(B)

P(B/ A) 0 P(B) P(A)

Remark: A S then P(A) P(S)

P(S) 1, P(A) 0

then 0 P(A) P(S) 0 P(A) 1

12
Theorem 4: P(A/ B) P(A)  P( A  B)

Proof: A (A/ B)  (A  B) since (A/ B)  (A  B) 

P(A) P (A/ B)  (A  B)  P(A/ B)  P(A  B)

P(A/ B) P(A)  P( A  B)

Theorem 5: If A and B are any event then

P(A  B) P(A)  P(B)  P(A  B)

Proof: A  B (A/ B)  B
By using Theorem 4

P(A  B) P(A/ B)  P(B) P(A)  P(A  B)  P(B)

P(A)  P(B)  P(A  B)
13
14
Theorem 6: If A1, A2,… An are any event then;
n
P(A1  A 2  ...  A n )  P(A i )   P(A i  A j ) 
i 1 1i  jn


1i  j k n
P(A i  A j  A k )  .......  (  1) n  1 P(A1  A 2 ...  A n )

BOOLE INEQUALITY: P(A1  A 2 ) P(A1 )  P(A 2 )

 n  n
If A1, A2,… An are any event P  A i   P(A i )
 i 1  i 1

BONFERONNI INEQUALITY P ( A1  A2 ) P ( A1 )  P( A2 )  1

 n  n
General form of this inequality P  A i   P(A i )  1
 i 1  i 1
Bonferroni's inequality is used to obtain a lower limit where it is impossible to
calculate the probability of intersection. 15
Solution: P ( A  B )  P ( A)  P (B)  1 0.90  0.95  1 0.85

Theorem 7: If A and B are any events, than

P(A) P(A  B)  P(A  B)

Example: A card is randomly selected from a deck of 52 cards. Let;

A The card is Spades Find P(A), P(B) and
B The card is colorful ( Jack , Queen or King ) P( A  B)

Note: Jack  Bacak

Hearts  Kupa
Queen  Kız Spades  Maça
King  Papaz Diamonds  Karo
Ace  As C lub s  Sinek
Solution:

16
Example: Two items are randomly chosen from a set of 12 items of which
4 are defective.

A  Both of the chosen items are defective

Find P(A), P(B) and
B  Neither of the chosen items are defective
P(C).
C  At least one of the chosen items are defective

Solution:

17
Example: Points a and b are randomly selected on a R real line as shown
below where  2 b 0 and 0 a 3. What is the probability that
the distance d between a and b is longer than 3?

Solution:

18
Example: The probability that a student passes from mathematics is 4/5,
the probability that a student passes from physics is 2/3 and the probability
that the student passes from both subjects is ½. Find the probability that
the student
a)Fails physics,
b)Passes mathematics or physics,
c)Passes mathematics or fails physics.

Solution:

19
Example: If two dies are rolled, what is the probability that the total number
comes up 9?

Solution:

20
Example: A coin is tossed three times.
A  Comes up head at least two times
B  Comes up head at the sec ond tos sin g 
C  Comes up head only at the sec ond tos sin g 
a)Are A and B disjoint events,
b)Are A and C disjoint events,
c)Are B and C disjoint events,
d)Find P ( A  B ), P ( A  C ), P ( B  C )

Solution:

21
CONDITIONAL PROBABILITY:
Let A and B are two events defined on a sample space S. Then, the
probability of the event A occuring when it is known that the event B has
occured is called a conditional probability and it is denoted by
P(A B)

The symbol P(A B) is usually read «the probablity that A occurs given
that B occurs» or simply «the probability of A, given B». We can consider

P(A) P(A S)

Thus, B is said to be a reduced sample space for this situation.

Definition: The conditional probability of A, given B denoted by P(A B)

is defined by

P(A B)
P(A B)  , P(B)  0
P(B)
22
Example: The people in a small town are categorized as shown in the
table:
Employed (E) Unemployed
(U)
Male (M) 300 400
Female (F) 250 250

Find P(U M), P(M U) ?

Example: It is known that the outcomes is an even number when a

balanced die is tossed. Accordingly, what is the probability that this number
is 4?

23
Example: 100 people are surveyed in order to determine relationship
between high tension and obesity. The results are given as:
Thin Normal Fat Total
High 8 10 22 40
Tension
Non-high 24 16 20 60
Tension
P(U M), P(M U) ?
Total 32 26 42 100

Find the probability that a randomly selected person

a)is fat,
b)has high tension,
c)is fat given that he/she has high tension
d)has high tension given that he/she is fat.

24
Example: Let two dies be rolled. What is the probability that both of them
come up 4 given that they comes up 8 in total.
Solution:

Example: Two balls successively selected without replacing them back,

from a bag that contains 3 green balls and 5 red balls. What is the
probability that the second chosen ball is green given that the first ball is
green?
Solution:

25
 MULTIPLICATION RULE FOR CONDITIONAL PROBABILITY:
* The probability of occurrence of the intersection of two events can be
calculated using the product rule.
P(A1  A 2 )
P(A1 A 2 )  P(A1  A 2 ) P(A1 A 2 ).P(A 2 )
P(A 2 )
P(A 2 )  0
or
P(A1  A 2 )
P(A 2 A1 )  P(A1  A 2 ) P(A 2 A1 ).P(A1 )
P(A1 )
P(A1 )  0

*In the case of three events;

P(A1  A 2  A 3 )
P(A 3 A1  A 2 ) 
P(A1  A 2 )
P(A1  A 2  A 3 ) P(A1  A 2 )P(A 3 A1  A 2 )

P(A1  A 2  A 3 ) P(A1 )P(A 2 A1 )P(A 3 A1  A 2 )

*If this formula is generalized for n events A1, A2,… An ,

P(A1  A 2  ...  A n ) P(A1 ).P(A 2 A1 ).P(A 3 A1  A 2 )....P(A n A1  A 2  ...A n  1 )
 n    n 1  
P  A i  P(A1 ).P(A 2 A1 ).P( A 3 A1  A 2 )....P  A n  Ai  
 i 1    i 1   26
Example: 4 bulbs are defective in a box of 18 bulbs. 2 bulbs are
successively selected without replacing them back. What is the probability
that
a)Both bulbs are defective,
b)Both bulbs are non-defective,
c)The first chosen bulb is defective and the second one is non-defective,
d)The first chosen bulb is non-defective and the other one is defective,

Solution:

Example: 4 bulbs are defective in a box of 15 bulbs. 3 bulbs are

successively selected without replacing them back. What is the probability
that all bulbs are non-defective?
Solution:

27
Example: A box contains 8 blue balls and 6 red balls. 3 balls are selected
successively from the box. Find the probability that the first ball is red, the
second ball is blue and third one is red, if 3 balls are chosen
a)Without replacing it back,
b)Replacing them back.

Solution:

28
DISCRETIZATION OF SAMPLE SPACE:

Let Bi  S, a) P(Bi  B j )  i j
n
b) B i S
i 1

c) P(Bi )  0 for all i

then B1, B2,… Bn events show a discretization of the sample space S.

The figure shows discretization of S for n = 8.

29
Theorem: According to the definition given above,

P(B1 )  P(B2 )  ...  P(Bn ) 1

Theorem: If Bi (i=1,2,…,n) is a disretization of a sample space S, then

n n
P(A)  P(Bi  A)  P (Bi ).P(A Bi )
i 1 i 1

for any events A in S.

Proof:

A A  B1   A  B2   ...  A  Bn 
P(A) P A  B1   P A  B2   ...  P A  Bn  Since A  Bi
P(A Bi ) P(Bi )P(A Bi ) are disjoint events

P(A) P(B1 )P(A B1 )  ......  P(Bn )P(A B n )

30
Example: A type of machine is produced in the factories A, B and C. The
number of machines produced in A is 2 times of machines produced in B
and also in C. It is known that 2% of machines produced in A as well as in
B, and 4% of machines produced in C are defective. If a machine is
randomly selected from the store in which all produced machines put, what
is the probability that the machine is defective?

Solution:

31