STATISTICS BOOTCAMP
SHEKHAR TOMAR
Assistant Professor
Indian School of Business
1
Random
2.1 variables
Set Theory
Definitions
• Experiment - Any observation or measurement of a
random phenomenon
• Basic outcome – a possible outcome of a random
experiment. Basic outcomes are mutually exclusive
and collectively exhaustive
3
Definitions
• Sample Space - The collection of all possible
outcomes of a random experiment.
• Event – Any subset of basic outcomes from the sample
space
4
example
Tossing a Coin
Sample space = {H,T}
Basic outcomes = {H}, {T}
Event: Getting a tail
5
example
Rolling a Die
Sample space = {1,2,3,4,5,6}
Basic outcomes = {1}, {2}, {3}, {4}, {5}, {6}
Event: Getting an odd number = {1, 3, 5}
6
Definitions
Complement - an event opposite to the event of
interest is called a complementary event
If A is the event => A’ is the complement (or AC ,A)
The sum of probability of events A and A’ is equal to 1
P(A) + P(A’) = 1 Or P(A) = 1 - P(A’)
7
Mutually Exclusive
If A and B are mutually exclusive:
• A can happen or
• B can happen,
• but both can not happen at the same time, then
P(A) + P(B) = 1
If A and B are not mutually exclusive:
• both can happen at the same time
P(A) + P(B) ≠ 1
8
Set Theory
A set is a collection of objects, which are the elements
of the set.
• If S is a set and x is an element of S, we write 𝒙 ∈ 𝑺.
• If x is not an element of S, we write 𝒙 ∉ 𝑺.
A set can have no elements, in which case it is called
the empty set, denoted by ∅.
9
Set Theory
Sets can be specified in a variety of ways
• If S contains a finite number of elements, say x1, x2,
. . . , xn, we write it as:
𝑺 = 𝒙𝟏 , 𝒙𝟐 , … . . , 𝒙𝒏
For example, the set of all possible outcomes of a die
roll is {1, 2, 3, 4, 5, 6}
10
Basic Set Operations
The complement of subset A is denoted by 𝑨
ഥ , 𝑨𝒄 𝒐𝒓 𝑨′
The intersection of A and B is denoted by 𝑨 ∩ 𝑩
The union of A and B is denoted by 𝐀 ∪ 𝐁
11
Basic Set Operations
A and B are called mutually exclusive or disjoint if
𝑨 ∩ 𝑩 = ∅ where ∅ is the empty set
If A1, A2, A3......An are subsets of a sample space S , then
they are called collectively exhaustive if
𝑨𝟏 ∪ 𝑨𝟐 ∪ 𝑨𝟑 ∪…………∪ 𝑨𝒏 = 𝑺
12
Random
2.2 variables
Venn Diagrams
Venn Diagrams
These are an excellent way of representing a
Probability Space
We can use them to clearly represent a situation and to
calculate corresponding probabilities
14
Venn Diagrams
Event: Even Number on a roll of Dice
A’
{1}, {3}, {5} A
{2}, {4}, {6}
15
Union
X = {1,2,3} and Y = {2,4,5,6}
X ∪ Y = {1,2,3,4,5,6}
X Y
16
Intersection
X = {1,2,3} and Y = {2,4,5,6}
X ∩ Y = {2}
X Y
17
Mutually Exclusive
X = {1,2,3} and Y = {4,5,6}
X∩Y= ∅
X Y
18
Probability Calculation
X Y X Y
P(X) = Pink space P(Y) = Blue space
/Total Space /Total Space
19
Probability Calculation
X Y
P(X ∪ Y) = (Pink + Blue space) P(X ∩ Y) = Green space
÷ Total Space ÷ Total Space
20
Basic Set Operations
A A
A B B
B
Shaded Region Shaded Region Shaded Region
is 𝐴 ∩ 𝐵 is 𝐴 ∪ 𝐵 is 𝐴 ∪ 𝐵𝑐
A A B
B
B C A C
Here 𝐵 ⊂ A. A, B, C are A, B, C are
Shaded disjoint partition of
Region is Ac set Ω 21
Example-I
The manager of a factory claims that among his 400
employees:
• 312 got a pay rise
• 248 got increased pension benefits
• 173 got both
• 13 got neither
22
Example-I
Using last years figures as your guide to this years
prospects, calculate the probability of:
• Getting a pay rise
• Not getting a pay rise
• Getting both a pay rise and pension benefits
23
Example-II
Of all flashlights in a large shipment, 15% have a
defective bulb, 10% have a defective battery, and 5%
have both defects. If you purchase a flashlight from the
shipment what are the probabilities of the following:
• A defective light bulb or a defective battery
• A good bulb or a good battery
• A good bulb and a good battery
24
Random
2.3 variables
De Morgan’s Law
De Morgan’s Law
If A and B are events and subsets of a universal set Ω
(usually the set of outcomes of some experiment) then,
ഥ∩ 𝑩
𝐀 ∪𝑩= 𝑨 ഥ
ഥ∪ 𝑩
𝐀 ∩𝑩= 𝑨 ഥ
26
De Morgan’s Law
27
De Morgan’s Law
ഥ∩ 𝑩
𝐀 ∪𝑩= 𝑨 ഥ
ഥ∪ 𝑩
𝐀 ∩𝑩= 𝑨 ഥ
Given the above, below also holds true:
ഥ∩𝑩
𝐏(𝐀 ∪ 𝑩) = 𝐏(𝑨 ഥ)
ഥ∪𝑩
𝐏(𝐀 ∩ 𝑩) = 𝐏(𝑨 ഥ)
28
Random
2.4 variables
Conditional
Probability
Conditional probability: P(A|B)
Probability of event A occurring given that event B
has occurred
P(A|B) = P(A∩B)/ P(B)
A B
30
example
Employees joining a firm in recent years
Sales Finance Operations Total
Female 616 194 26 836
Male 529 171 30 730
Total 1145 365 56 1566
𝟕𝟑𝟎 𝟑𝟎
𝑷(𝑴𝒂𝒍𝒆) = ≅ 𝟎. 𝟒𝟔 𝑷 𝑴𝒂𝒍𝒆 𝑶𝒑𝒆𝒓𝒂𝒕𝒊𝒐𝒏𝒔 = ≅ 𝟎. 𝟓𝟑
𝟏𝟓𝟔𝟔 𝟓𝟔
31
Contingency table
Joint
B B’
A P(A ∩ B) P(A ∩ B) P(A)
A’ P(A’ ∩ B) P(A’ ∩ B’) P(A’)
P(B) P(B’) P(S)=1
Marginal
32
Conditional probability
𝑃 𝐴∩𝐵
𝑃 𝐴𝐵 =
𝑃 𝐵
Conditional probability can be rewritten as follows:
𝑃 𝐴 ∩ 𝐵 = 𝑃 𝐴 𝐵 × 𝑃(𝐵)
What’s the use of this ?
33
Law of Total Probability
• The Law of Total Probability can be extended to
more than two mutually exclusive and collectively
exhaustive events
• If B1, B2, …, Bn are mutually exclusive and
collectively exhaustive events, then the probability
of an event A an be written as
𝑃 𝐴 = 𝑃 𝐴 ∩ 𝐵1 + 𝑃 𝐴 ∩ 𝐵2 + … . . +𝑃 𝐴 ∩ 𝐵𝑛
34
example
You enter a chess tournament where your probability
of winning a game is:
• 0.3 against half the players (call then type 1),
• 0.4 against a quarter of the players (call then type 2),
• and 0.5 against remaining quarter of the players (call
then type 3).
You play game against a randomly chosen opponent.
What is the probability of winning?
35
example
• Let Bi = event of playing opponent of type i.
P(B1) = 0.5, P(B2) = 0.25, P(B3) = 0.25
Also let A = event of winning, we have
P(A|B1) = 0.3, P(A|B2) = 0.4, P(A|B3) = 0.5
Now use Law of Total Probability to find P(A).
36
independence
• Two events A and B are said to be independent if:
P(A|B)=P(A) & P(B|A)=P(B)
• INDEPENDENT EVENTS: The occurrence of one
event has no effect on the probability of the other.
37
example
Red Black Total
Ace 2 2 4
Non-Ace 24 24 48
Total 26 26 52
P(Red U ACE )=?
38
Random
2.5 variables
Bayes’ Theorem
Breast cancer
Cancer (1%) No Cancer (99%)
Test (+ve) 80 9.6
Test (-ve) 20 90.4
What is the probability of cancer if a person
tests positive?
40
Breast cancer
Cancer (1%) No Cancer (99%)
Test (+ve) 80 9.6
Test (-ve) 20 90.4
Cancer (1%) No Cancer (99%)
Test (+ve) True Positive False Positive
0.008 0.095
Test (-ve) False Negative True Negative
0.002 0.895
41
Bayes’ theoreM
• Given two uncertain events H and E.
𝑷 𝑯∩𝑬 𝑷 𝑯 𝑷 𝑬𝑯
𝑷 𝑯𝑬 = =
𝑷 𝑬 𝑷 𝑬
𝑷 𝑯 𝑷 𝑬𝑯
=
𝑷 𝑯 𝑷 𝑬 𝑯 + 𝑷 𝑯′ 𝑷 𝑬 𝑯′
42