Quantitative Techniques

PROBABILITY AND STATISTICS

Semester - I

By Dr. Santosh Kumar Gupta

 Are these the same type of Probability
u There is a 50% chance that India will win the toss.

u My guess is answer ”a” is the right choice.

u There is a 30% chance of rain tomorrow.

u The probability of scoring on a penalty shot in ice hockey is 0.47.

u The probability that the current Mayor will resign is 0.85.

u The probability of rolling two sixes with two dice is 1/36.

u The probability that you will go Europe this year is 0.14.

 Probability
uAs a general rule, to be able to draw valid
inferences about a population from a sample,
one needs to know how likely it is that certain
events will occur under various circumstances.

uThe determination of the likelihood, or chance,

that an event will occur is the subject matter of
probability.
 Basic Probability Concepts
u Experiment - In statistics, any process or phenomenon
that we wish to study statistically is called an experiment.
u In other words, when we want to study statistically to see
if there are statistical patterns in the data or phenomena,
we use the word experiment to denote that.
u An experiment is any process that produces an
observation or outcome.
u Examples: Tossing a coin, Rolling a die.
uA random experiment is an experiment whose outcome
is not predictable with certainty.
 Basic Probability Concepts
u Outcome - The result of an experiment is referred to
as the Outcome of that experiment. It may also be
referred to as data in some sense.
uSample Space - The sample space is a set that
contains all outcomes of an experiment. It is
typically denoted by ‘S’ or ‘Ω’.
uEvent – An event is a collection of basic outcomes.
An event is a subset of the sample space. An event
is said to have occurred if the outcome is contained
in the subset.
Basic Probability Concepts – Sample Space
Sample Space and Events
 Union of Events
u Union of Events - For any two events A and B , we define
the new event A U B called the union of events A and B , to
consist of all outcomes that are in A or in B or in both A and
B.
u The event A U B will occur if either A or B occurs.

u Experiment: Tossing two coins and noting the outcomes

u Event:
uSampleSpace S = {HH, HT, TH, TT}
uHead on the first toss A1 = {HH, HT}
uHead on second toss A2 = {HH, TH}
uHead on first or second toss A1 U A2 = {HH, HT , TH}
 Intersection of Events
u Intersectionof Events - For any two events A and B , we
define the new event A ⋂ B called the intersection of
events A and B , to consist of all outcomes that are in A
and in B.
u The event A ⋂ B will occur if both A and B occurs.

u Experiment: Tossing two coins and noting the outcomes

u Event:
uSample Space S = {HH, HT, TH, TT}
uHead on the first toss A1 = {HH, HT}
uHead on second toss A2 = {HH, TH}
uHead on first or second toss A1 ⋂ A2 = {HH}
Union and Intersection of Events
 Null and Disjoint Events
uNull Events – An event without any outcome is
called a null event and is denoted by ɸ.

uDisjoint Events – If the intersection of A and B

is the null event, and since A and B cannot
simultaneously occur, we say that A and B are
disjoint or mutually exclusive.
Mutually Exclusive or Disjoint Events
 Compliment of an Event
uCompliment of an Events – Compliment of an
event A is denoted by Ac and consists of all the
outcomes in the sample space S that are not in A.
uAc will occur if and only if A does not occur.
u Experiment: Tossing two coins and noting the outcomes
u Event:
uSample Space S = {HH, HT, TH, TT}
uHead on the first toss A1 = {HH, HT}
uHead on second toss A2 = {HH, TH}
uA1c = {TH, TT} and A2c = {HT, TT}
Compliment of an Event
 Subset of an Event
u Subset of an Events – For any two events A and B, if
all the outcomes in A are also in B, we say that A is
contained in B, or A is a subset of B, and it is
denoted by, A ⊂ B.

u Experiment: Tossing two coins and noting the outcomes

uEvent:
uSample Space S = {HH, HT, TH, TT}
uHead on the first toss A1 = {HH, HT}
uHead on second toss A2 = {HH, TH}
uA ⊂ B = {HH}
 Collectively Exhaustive Event

u Collectively Exhaustive Events – Events are

collectively exhaustive, if at least one of the events
must occur when an experiment is conducted.
u Partitions – If two or more disjoint events make up
the whole sample space together, then it is referred
to as the Partition.
uThe event and its complement are always disjoint events,
i.e. A ∩ Ac = empty set
uThe event and its complement together cover all the
outcomes, i.e. A ∪ Ac = S
Collectively Exhaustive Event
Set Operations
De-Morgan’s Law
 Types of Probabilities - Classical
uClassical Definition: If S be the sample space of
a random experiment in which there are n
equally likely outcomes, and the event A
consists of exactly m of these outcomes, then
we say the probability of the event A is m and
n represent it as P(A ) = m / n.
u Classical
prob is also called a priori probability
because we don’t need to perform experiments.
 Types of Probabilities - Empirical
uEmpirical
Probability (Aposteriori or Relative
Frequency): The probability of an event in an
experiment is the proportion (or fraction) of times the
event occurs in a very long (theoretically infinite) series of
(independent) repetitions of experiment. In other words,
if n(A ) is the number of times A occurs in n repetitions of
the experiment, P(A ) = n(A ) / n

u Probability of an Event (A) =

u=Number of Times the event A occurs in repeated trials /
Total number of trials in a random experiment
 Types of Probabilities - Subjective
uSubjective Probability: The probability of an
event is a “best guess” by a person making
the statement of the chances that the event
will happen. The probability measures an
individual’s degree of belief in the event.
uBecause most higher-level social and managerial
decisions are concerned with specific, unique
situations, rather than with a long series of identical
situation, decisions makers often use this
probability.
 Types of Probabilities
uClassical Probability: Based on
Mathematical Formula

uEmpiricalProbability: Based on relative

frequencies of historical data

uSubjective
Probability: One-Shot
educated guess
Basic Properties of Probability (Axioms)
uRule 1: The probability of an outcome in
the sample space is 1, P(S) = 1

uRule 2: For any event A, the probability of

A is between 0 and 1, 0 ≤ P(A) ≤ 1.

uRule3: For disjoint or mutually exclusive

events A and B,
uP(A U B) = P (A OR B) = P(A) + P(B)
 General Properties of Probabilities
u1.Probability of complement of an event:
uP(Ac) = 1− P(A)
uA and Ac are disjoint. Also, A U Ac = S
uApply Property 3 to LHS P(A U Ac ) = P(A) + P(Ac)
uApply Property 2 to RHS P(S) = 1
uEquating both, we get
uP(A U Ac ) = P(A) + P(Ac) = P(S) = 1.
uHence P(Ac ) = 1− P(A)
 General Properties of Probabilities
u2.P(Φ) = 0
uSc = Φ
uApply the above property, P(Sc) = 1− P(S)
uApply property 2, P(S) = 1
uHence, P(Φ) = 0
 Addition Rule of Probability
uThe addition rule of probability relates the
probability of the union of events. If A and B
are two events, which are not necessarily
disjoint, then union of P(A) and P(B), that is P(A
U B) is calculated using the addition rule of
probability.

uFor any events A and B,

uP(A U B) = P(A) + P(B) − P(A ∩ B)
 Addition Rule of Probability
u Let
S be a sample space when rolling a fair Dice. Let A be an
event where outcome is odd numbers and B be the event
where outcome is greater than 4. Find the probability of event
C that the outcome is either odd or greater than 4. (0.67)

uA certain person encounters three traffic lights when driving to

work. Suppose that the following represent the probabilities of
the total number of red lights that she has to stop for: P (0 red
lights)= 0.14; P (1 red light)= 0.36; P (2 red lights)= 0.34; P (3 red
lights)= 0.16
u (a). What is the probability that she stops for at least one red light
when driving to work?
u (b). What is the probability that she stops for more than two red lights?
 Addition Rule of Probability
u A customer that goes to the clothing store will purchase a shirt with
probability 0.3. The customer will purchase a pant with probability
0.2 and will purchase both a shirt and a pant with probability 0.1.
What proportion of customers purchases:
u (a). either a shirt or a pant? (0.4)
u (b). neither a shirt nor a pant? (0.6)

u A student has a 40 percent chance of receiving an A grade in

statistics, a 60 percent chance of receiving an A in mathematics,
and an 86 percent chance of receiving an A in either statistics or
mathematics. Find the probability that she
u (a). Does not receive an A in either statistics or mathematics. (0.14)
u (b). Receives A’s in both statistics and mathematics. (0.14)
 Addition Rule of Probability
u When drawing a card from standard deck of 52 playing cards,
what is the probability that the card is either red or a queen. (0.54)
 Addition Rule of Probability
u An inspector of the Alaska pipeline has the task of comparing
the reliability of two pumping stations. Each station is
susceptible to two kinds of failure: pump failure and leakage.
When either (or both) occur, the station must be shut down. For
Station one, probability of Pump Failure is 0.07, while for Station
two, it is 0.09. Probability of leakages for station one is 0.10 while
for station two it is 0.12. If the probability of occurring either
pump failure or Leakages for station two is 0.06 and for station
one, it is nil, find which station has the higher probability of
being shut down. (0.17; 0.15).
u Two cards are randomly selected from a deck of 52 playing
cards. What is the conditional probability they are both aces
given that they are of different suits? (0.006)
 Empirical Probability
uStatisticalIndependence: If the occurrence of one
event has no effect on the occurrence of any other
event. Ex: Tossing of two fair coins.

uJointProbabilities: Probabilities of two or more

independent events occurring together.

uMarginal Probability: Probability of observing and

outcome with a single attribute, regardless of any
other attribute.
 Empirical Probability
u Joint Probabilities:
Probabilities of
intersection: Probabilities
of Female and owning
phone is 0.34; Male and
not owning phone is 0.14.
etc.

u Marginal Probabilities:
Probabilities at the
margins of the
contingency table:
Probability of Female =
(0.10 + 0.34 = 0.44);
owning phone – (0.34
+0.42 = 0.76) etc.
 Empirical Probability
u Conditional Probability:
The probability of event B
given that event A has
occurred.
u P (B | A)
u What is the probability
that someone owns a
phone given that buyer is
a female.
u What is the probability
someone is a male given
that s/he does not own a
phone.
 Conditional Probability

u What is the probability that someone owns a phone given

that buyer is a female.

u Whatis the probability someone is a male given that s/he

does not own a phone.
 Conditional Probability
u When rolling a fair dice
twice, A is an event of
first roll lands on 4, while
B is the event that sum of
pair of rolled dice is 10.
What is the probability
that sum of pair of rolled
dice is 10 given that the
first roll lands on 4.
u P(A | B) = P(A ⋂ B) / P(B)
 Conditional Probability
u Ifthe probability of occurrence of event B is affected by
the occurrence of event A, then event A and B are
statistically dependent.
u P(A | B) = P(A ⋂ B) / P(B)
u P(A ⋂ B) = P(B) x P(A|B)

u When rolling a fair dice twice, A is an event of first roll

lands on 4, while B is the event that sum of pair of rolled
dice is 10. What is the probability that sum of pair of rolled
dice is 10 given that the first roll lands on 4.
 Conditional Probability – Statistical Dependence

u According to a survey, the probability that a family owns two

cars if its annual income is greater that Rs 35000 is 0.75. Of the
household surveyed 60 % had income over Rs 35000 and 52 %
had two cars. What is the prob that a family has two cars and
an income over Rs 35000 a year? (0.45)
u In an introductory statistics class of forty students, the number
of males is equal to 23 and number of females is equal to 17.
Two students are selected at random from the class. The first
student selected is not returned to the class for possible
reselection; that is, the sampling is without replacement. Find
the probability that the first student selected is female and the
second is male. (0.251)
 Conditional Probability – Statistical Independence

u If
the probability of occurrence of event B is not
affected by the occurrence of event A, then event
A and B are statistically independent. P(B|A) = P(B).

u What is the probability that the second toss of a fair coin

will result in heads, given that heads resulted in first toss.
u What is the probability that in selecting two cards one at a
time from a deck with replacement, the second card is:
u (a) A face card, given that the first card was red? (0.23)
u (b) An ace, given that the first card was a face card? (0.076)
u (c) A black jack, given that the first card was a red ace? (0.038)
 Multiplication Rules of Probability
u IfA and B are independent events then,
uP(A ⋂ B) = P(A) x P(B)

uIf a dice is rolled twice, and let A be the event

that first outcome is 3, B be event that sum of
outcomes be 8, and C be the event that sum of
the outcomes be 7.
uAre events A and B Independent.
uAre the events A and C independent.
 Conditional Probability – Independence
u IfA and B are independent events then,
uP(A ⋂ B) = P(A) x P(B)

u Consider an experiment of randomly selecting one

card from a deck of 52 playing cards. Event A is a
face card selected, Event B is a King selected, Event C
is a heart selected.
uAre events A and B Independent.
uAre the events B and C independent.
Conditional Probability – In/dependence Caselet
 Conditional Probability - Independence
u If A and B are independent events then,
uA and Bc are also independent.
uA = (A ⋂ B) U (A ⋂ Bc)

u Since (A ⋂ B) and (A ⋂ Bc) are

Mutually exclusive and independent
uP(A) = P(A ⋂ B) + P(A ⋂ Bc)
u P(A) = P(A) x P(B) + P(A ⋂ Bc)
 Conditional Probability - Independence
uP(A) = P(A ⋂ B) + P(A ⋂ Bc)
u P(A) = P(A) x P(B) + P(A ⋂ Bc)
u P(A) x (1-P(B)) = P(A ⋂ Bc)
u P(A ⋂ Bc) = P(A) x P(Bc)
u If A is independent of B, then
probability of A’s occurrence is
unchanged by information
whether or not B has occurred.
 Conditional Probability - Independence
u P(A)= P(A ⋂ B) + P(A ⋂ Bc)
u P(A ⋂ B) = P(B) x P(A | B)
u P(A ⋂ Bc) = P(Bc) x P(A | Bc)
u P(A) = P(B)*P(A|B) + P(Bc)*P(A|Bc)

u Probabilityof event A is a weighted

average of the conditional
probability of A given that B occurs
and the conditional probability of A
given that B does not occur.
 Conditional Probability - Independence
u An insurance company believes that people can be
divided into two classes—those who are prone to
have accidents and those who are not. The data
indicate that an accident-prone person will have an
accident in a 1-year period with probability 0.1; the
probability for all others is 0.05. Suppose that the
probability is 0.2 that a new policyholder is
accident-prone. What is the probability that a new
policyholder will have an accident in the first year?
(0.06).
 Conditional Probability - Independence
u A : A new policy holder will have an accident in the first year.
u B : A new policy holder is accident prone.
u Given an accident-prone person will have an accident in a 1-year
period with probability 0.1, i.e., P(A|B ) = 0.1
u The probability for all others is 0.05, i.e., P(A|Bc ) = 0.05
u The probability that a new policyholder is accident-prone = P(B) =
0.2
u The probability that a new policyholder will have an accident in the
first year, P(A).
u P(A) = P(B) x P(A|B ) + P(Bc) x P(A|Bc)
u 0.2 ×0.1 + 0.8 × 0.05 = 0.06
 Bayes’ Rule

uP(B|A) = (P(A|B) x P(B)) / P(A)

u= P(A|B) x P(B) / P(B)*P(A|B) + P(Bc)*P(A|Bc)
 Bayes’ Rule
uAn insurance company believes that people can
be divided into two classes—those who are prone
to have accidents and those who are not. The
data indicate that an accident-prone person will
have an accident in a 1-year period with
probability 0.1; the probability for all others is 0.05.
Suppose that the probability is 0.2 that a new
policyholder is accident-prone. If a new
policyholder has an accident in the first year, what
is the probability that he or she is accident-prone?
(0.333)
 Bayes’ Rule
u The probability that a new policyholder will have an accident in the
first year, P(A).
u P(A) = P(B) x P(A|B ) + P(Bc) x P(A|Bc) = 0.2 ×0.1 + 0.8 × 0.05 = 0.06
u Now “If a new policyholder has an accident in the first year”,
implies occurrence of event A
u What is the probability that he or she is accident-prone? In other
words, what is
u P(accident prone|policy holder has an accident in first year).
u This is equivalent to P(B|A). Applying Bayes’ rue we get
u P(B|A) = P(A|B) x P(B) / P(B)*P(A|B) + P(Bc)*P(A|Bc) = 0.02 / 0.06 =
0.333
 Bayes’ Rule
u Ina factory that produces bolts, three machines—A,
B, and C—are responsible for manufacturing 25%,
35%, and 40% of the total bolts, respectively. The
defect rates of these machines differ: machine A
produces 5% defective bolts, machine B produces
4% defective bolts, and machine C produces 2%
defective bolts. A bolt is randomly selected from the
overall production, and it is found to be defective.
What is the probability/likelihood that this defective
bolt was produced by machine B? (0.406)
 Bayes’ Rule
uIna test an examinee either guesses, copies or
knows the answer to a multiple-choice
question with four choices. The probability that
he makes a guess is1/3 and the probability that
he copies the answer is 1/6. Given that the
probability of his answer being correct when
he copied it is 1/8, what is the probability that
he actually knew the answer to the question,
considering the probability that his answer is
correct if he copied it? (0.827)
Subjective Probability (Case-let)
Subjective Probability – Conjunction Fallacy
 Benford’s Law
u The first digits of the numbers found in series of
records of the most varied sources do not display a
uniform distribution, but rather, they’re arranged in
such a way that the digit one is the most frequent,
followed by two, three, and so in a successively
decreasing manner down to nine.

u This
law can be utilized to detect patterns, or lack
thereof, in naturally occurring data sets, which can
be used to help catch anomalies or fraud in data.
Benford’s Law