DISC: 203 – PROBABILITY &

STATISTICS

PROBABILITY

1 Lecturer: Muhammad Asim

PROBABILITY

2
 DEFINITIONS

 An experiment is a process which leads to a single outcome (called

a sample point) that cannot be predicted with certainty.
 Example: We roll a dice once and observe the value that is faced up.

 A sample point is the most basic outcome of an experiment.

Collection of all sample points in an experiment is called sample
space.
 Sample space for the example: S = {1,2,3,4,5,6}

3
 DEFINITIONS
 The probability of a sample point is a number between 0 and 1 that
measures the likelihood that the outcome will occur when the
experiment is performed.
 Law of Large Numbers : When an experiment is replicated for a large
number of times, the relative frequency of an outcome approaches the
theoretical probability of the outcome.

4
 DEFINITIONS
 Probability rules for sample points:

 All sample point probabilities must be between 0 and 1

 The sum of all sample point probabilities within a sample space must be 1

5
DEFINITIONS - EVENT
 An event is a specific collection of sample points.

 Example: One roll of a fair dice. Define three events:

 A: {Observe an even number}
 B: {Observe a number less than or equal to 3 }
 C: {Observe 1}
 S = {1,2,3,4,5,6}

 P(A) = 1/2; P(B) = 1/2; P(C) = 1/6

 If all sample points are equally likely,

6
P(X) = # of elements of X / total # of sample points
EXAMPLE

7
DEFINITIONS - UNION

 The union to two events A and B occurs if either A or B or

both occur. We denote the union as A U B.

 In our example, roll a fair dice

 A = {observe an even number}

 B= {observe a number less than

or equal to 3}
 AUB = {1,2,3,4,6}

 P(AUB) = 1/6 +1/6+1/6+1/6+1/6 =5/6

8
DEFINITIONS - INTERSECTION
 The intersection to two events A and B occurs if both A and B
occur. We denote the intersection as A ∩ B.

 In our example, roll a fair dice

 A = {observe an even number}

 B= {observe a number less than

or equal to 3}
 A ∩ B = {2}

 P(A ∩ B) =1/6
9
DEFINITIONS- MUTUALLY EXCLUSIVE
EVENTS
 Events A and C are mutually exclusive if A ∩ C contains no
sample points that is, if A and C have no sample points in
common.
 In our fair dice example,
 A: {Observe an even number} & C: {Observe 1} are
mutually exclusive.

S
 P(A U C) = P(A) + P(C) 3
= 3/6 + 1/6 = 2/3 2 A C
4 6 1

P(A ∩ C) = 0 10
5
P(A) = 1/2
DEFINITIONS - COMPLEMENTARY
EVENTS
 The complement of an event A is event that A does not
occur. i.e. all sample points that are not in A. It is
denoted as Ac.
 Sum of the probabilities of A and Ac is equal to 1

 In our fair dice example,

 S= {1,2,3,4,5,6}

 C= {1}

 Cc = {2,3,4,5,6}

 P(C) + P(Cc ) =1

11
ADDITIVE RULE OF PROBABILITY
 P(A U B) = P(A)+ P(B) – P(A ∩ B)

 In our example, roll a fair dice

 A = {observe an even number}

 B= {observe a number less than

or equal to 3}

 P(A U B) = P(A)+ P(B) – P(A ∩ B)

= 3/6 + 3/6 – 1/6 = 5/6
12
CONDITIONAL PROBABILITY

 Suppose we roll a fair dice. Let E={observe a six)

 We know P(E) = 1/6

 If someone secretly rolls a dice and tells you she has witnessed
event A = {observe an even number}
 Now, given event A, what are the chances that event E, a six was
observed? 1/3 [because sample space was reduced to A]

 We call this the conditional probability of E given A

 Mathematically, P(E / A) = P(E ∩A) / P(A) 13

= (1/6) / (3/6) = 1/3
•
EXAMPLE
 A consumer equipment manufacturer conducted the analysis
of consumer quality complaints. All complaints fell into six
categories.
 What is the probability that a received customer complaint
that originated during product guarantee period was caused
by product appearance?
 0.32/(.18 + .13 +.32) = 0.51

Reason for Complaint

Electrical Mechanical Appearance
During Guarantee Period 18% 13% 32%
After Guarantee Period 12% 22% 3%
14
THE MULTIPLICATIVE RULE
 P(A ∩ B) = P(A|B) × P(B) = P(B|A) × P(A)

 Example: An in investor in wheat futures is concerned

with the following events
 A: Production of wheat is profitable next year

 B: A serious drought occurs next year

 Investor knows that P(A|B)= 0.01 and P(B) = 0.05

 What is the probability that an investment is profitable

while a serious drought also occurs?
 P(A ∩ B) = P(A|B) × P(B) = (0.05)(0.01) = 0.0005
15
EXAMPLE

16
EXAMPLE 3.22 (FROM TEXT BOOK)

17
18
BAYES’ RULE
 Generally,let S , …, S be k mutually exclusive and
1 k

exhaustive events and let E be a particular outcome

of an experiment. Then,
P( Si  E ) P( Si ) P( E / Si )
P( Si / E )  
P( E ) P( E )
where
P ( E )  P( S1  E )  P( S 2  E )  ....  P ( S k  E )
 P( S1 ) P( E / S1 )  P( S 2 ) P( E / S 2 )  ....  P( S k ) P( E / S k )
19
INDEPENDENT EVENTS
 Events A and B are independent if the occurrence of B does not
alter the probability that A occurs
 P(A|B) = P(A)

 P(B|A) = P(B)

 Take fair dice and define

 A: Observe an even number  P(A) = ½

 B: Observe value ≤ 4  P(B) = ⅔

 Are they independent?

 P(A ∩ B) = ⅓

 P(A|B) = P(A ∩ B)/P(B) = ⅓ / ⅔ = ½ = P(A)

20
 Hence, independent
INDEPENDENT EVENTS
 If A and B are independent then

 P(A ∩ B) = P(A) P(B)

 The converse is also always true…

21
EXAMPLE
 Suppose a retail computer store owner is planning to
place an order for laptops. She is trying to decide how
many touch-screen laptops to order. The owner’s records
indicate that 80% of the previous customers purchased
regular and 20% purchased touch-screen laptops.
(a) What is the probability that the next two customers
will purchase touch-screen laptops? (0.2)(0.2)
(b) What is the probability that the next ten customers will
purchase touch-screen laptops? (0.2)^10

22
EXAMPLE
 A student has not prepared for the quiz and is randomly
guessing on all questions. The quiz has six questions,
each with four alternatives. What is the probability that
the student is able to guess five or more questions
correctly. Note: with random guessing, there is only a
0.25 probability of getting a question right. Assume the
questions are attempted independently.
 Hint: If x = # of questions guessed correctly

 P(x>=5) = P(x=5) + P(x=6)

 P(x=6) = (0.25)^6

 P(x=5) = # of ways 5 questions can be selected out

23
of the six * (0.25)^5 * (0.75)
COUNTING - COMBINATORIAL RULE

Use the combinations formula to determine the number of

ways to pick k elements out of available n.

n n!
C k
k! (n  k)!
• For the example, n=6, k=5
• So, the number of ways of answering five questions
correctly out of six are given by 6! / 5! = 6
• Answer to the example = 0.0046
24