Академический Документы
Профессиональный Документы
Культура Документы
Quantitative Techniques-I
Amitava Mukherjee;
Course Outline
Sl.
No.
Topic
Appox. No.
of Sessions
Faculty
Theory of Probability
Myself
Discrete Distributions
Myself
Continuous Distributions
Prof. S. K. De.
Elementary Notions of
Stochastic Processes
Prof. S. K. De.
Decision Theory
Prof. S. K. De.
Evaluation System
Component
Syllabus
Weight
Quiz I
Tentatively after 5
sessions
[Announced but
closed Book/Notes]
20%
Mid Term
Tentatively after
[Announced but
10 sessions
closed Book/Notes]
20%
Quiz-II
To be informed By To be informed By
Prof. S. K. De.
Prof. S. K. De.
20%
End Term
After all 20
sessions
40%
Time Line
Entire Course
Please refrain from asking a question that two of your friends have
already asked Rather meet personally and ask for more
clarifications
Grading
System
Rating
Fail
Grade
D+
C+
B+
A+
Quality
Points
** [10- [20- [34- [50- [75- [9020) 34) 50) 75) 90) 100]
Equivalent Academic
Percentiles Indiscipline
Poor
Average
Good
Excellent
Appointment
Always Welcome
Fix an Appointment by sending a Direct Email to me
No Voice Calling is entertained, in any form
Specially, Ill reserve Wednesday late evening to meet
students
Theory of Probability
Session-1
Degree of Uncertainty
Degree of Uncertainty
14
15
Definition of Probability
Various Aspects
Meaning of probability
As regards probability which expresses the uncertainty about the
observables, it is given radically different interpretations in the
objective and subjective approaches.
In the former, roughly speaking, we assume in effect that the
unpredictable variation of the evidence is such that the relative
frequency with which it would belong to any meaningful set in the
evidential domain would tend to stabilize around some idealized
value, if the number of repetitions were made indefinitely large.
The meaningful sets, technically called measurable sets, are those
which are of practical interest and are theoretically easy to handle
18
Meaning of probability
The basis of this assumption, which we call frequential regularity
is our experience with diverse types of particular repetitive
experiments.
This is commonly called statistical regularity [A misnomer]
For any set of interest the probability that the uncertain evidence
will belong to it is identified with the corresponding idealized
long-term relative frequency.
Probabilities, so defined, of all meaningful sets in the evidential
domain determine a probability distribution over the domain and
this gives an objective representation of the evidential uncertainty.
19
Meaning of probability
In the subjective approach probability exists only in one's mind and
may vary from person to person.
For a particular person the probability of any set of interest
represents the person's degree of belief in the materialization of the
event that the evidence (extended evidence) generated through the
operations when they are performed would belong to that set.
In practice this degree of belief can be quantified introspectively,
e.g. by ascertaining the maximum price one is prepared to pay
outright for a unit gain contingent on the actual realization of the
event.
20
Meaning of probability
Ideally one should attach numerical degrees of belief to
different sets of interest in a consistent or coherent manner.
Coherent probabilities for different meaningful sets in the
domain define a probability distribution over it.
Since uncertainty here means absence of knowledge, such a
probability distribution may cover evidence extended
backward or collaterally to involve unobserved characters
belonging to the past or the present.
21
23
28
31
33
Probability in Finance :
Consider a straightforward but rigorous framework for elaboration, with no
extraneous mathematical or philosophical baggage, of two ideas that are
fundamental to both probability and finance:
The Principle of Pricing by Dynamic Hedging : [Can be discerned in the
letters of Blaise Pascal to Pierre de Fermat in 1654] When simple gambles
can be combined over time to produce more complex gambles, prices for
the simple gambles determine prices for the more complex gambles.
The Hypothesis of the Impossibility of a Gambling System: Sometimes
we hypothesize that no system for selecting gambles from those offered to
us can both (1) be certain to avoid bankruptcy and (2) have a reasonable
chance of making us rich.
34
Probability in Marketing :
35
Theory of Probability
Session-2
Calculus of Probability :
Connections with Set Theory
Set Theory
Probability Theory
Notations
Element
Set
Universal Set
or
Null set
Impossible Event
Complement of a set A
Complementary event of A
is a subset of
( is a superset of )
Occurrences of events or
37
Calculus of Probability :
Connections with Set Theory
Set Theory
Probability Theory
38
Notations
Simple Examples
Consider a box with n white and m red balls. In this case,
there are two elementary outcomes: white ball or red ball.
Probability of selecting a white ball is
40
41
42
43
44
45
Travelers Choices
There are three major possible options available to a travel
agency for its customer who wants to travel to New Delhi
from Jamshedpur
Direct train to New Delhi
By Road/Rail to Ranchi and Flight from Ranchi
By Rail/road to Kolkata and Flight from Kolkata
Agency has records of previous bookings in the same route
over last few years that will help them to assess probable
choice of customers
46
Other Applications
Proportion of loan application rejected by a a major bank from
SMEs micro, small and medium-sized enterprises
Proportion of defective items produced by a manufacturing unit
In estimating life time of a product, proportion of electric bulb
survived after 1000 hours in operations.
In fact, estimating probability of survival of a
electric/electronic device after certain hours, we can actually
have a lifetime distribution that we study later.
Not only for consumer durables, we may think of Consumer
lifetime as well.
47
Combinatorics :
Arrangement of r balls in n cells
48
Combinatorics :
Arrangement of r balls in n cells
Exclusion principle
followed
(Maxwell-Boltzman
Statistics)
Balls are
indistinguishable
(Bose-Einstein Statistics)
(FermiDirac
Special Case:
statistics)
No cell remain empty:
Balls are
distinguishable
49
Application
A random sample of size with replacement is taken from a
population of elements. What is the probability that in the
sample, no element appears twice, that is, the sample could
have been realized also by sampling without replacement?
We see that there are possible sample in all of which
satisfies the stipulated condition. Assuming that all
arrangements have equal probability, we conclude that
probability of no repetition in our sample is
50
Industrial Implications
If in a coal mines, 12 accidents occur in each year, then
practically all year will contain months with two or more
accidents. The probability that all months will have one
accident each is only 0.0000537.
On the average, only one year out of 18614 years, will
show a uniform distribution of one accident per month
This example reveal an unexpected characteristic of pure
randomness
This type of argument is often used for fraud detection
51
Extensions
The number of ways to deposit distinct objects into cells
with objects in cell no. ( are non-negative integers
summing to )is
(ordering of bins is important but within each bin the
ordering is not important).
52
More Example
A throw of twelve dice can result in different outcomes
which we consider equally likely. The event that each face
appears twice can occur in as many ways as twelve dice can
be arranged in six groups of two each. The probability of
that event is therefore
53
Criticism of
Classical Definition of Probability :
Mathematicians find the definition to be circular.
The probability for a "fair" coin is... A "fair" coin is defined by a
probability of...
The definition is very limited. It says nothing about cases where no
physical symmetry exists.
Insurance premiums, for example, can only be rationally priced by
measured rates of loss.
It is not trivial to justify the principle of indifference except in the
simplest and most idealized of cases. Coins are not truly symmetric.
Can we assign equal probabilities to each side? Can we assign equal
probabilities to any real world experience?
56
V - algebra
A non-empty collection of subsets of Sample space is
called a sigma algebra (or Borel field for events over real
line), denoted by B), if it satisfied the following two
properties:
a. If , then
( is closed under complementation).
b. If , , , then
( is closed under countable unions).
To show that is closed under finite unions
To show ( the empty set is an element of ) and
( the sample space is an element of ).
57
V - algebra
Easy to realize that the superset generated by countable
sample space : is always a V - field.
Let : :. Can we
consider as a V - field?
Probability is a measure (set-function) defined on :
: is known as probabilizable space
58
More on V - algebra
Example-1. (Sigma algebra-I) If S is finite or countable, we
define for a given sample space S, B = {all subsets of S,
including S itself}. If S has n elements, there are sets in
B. For example, if S = {1, 2, 3}, then B is the following
collection of 2 3 = 8 sets: {1}, {1, 2}, {1, 2, 3}, {2}, {1, 3},
, {3}, {2, 3}.
Example-2. (Sigma algebra-II) Let S = (, ), the real line.
Then B is chosen to contain all sets of the form [a, b], (a,
b], (a, b), [a, b) for all real numbers a and b. Also, from the
properties of B, it follows that B contains all sets that can
be formed by taking (possibly countably infinite) unions
and interactions of sets of the above varieties.
59
Theory of Probability
Session-3
61
Andrey Kolmogorov
Andrey Nikolaevich Kolmogorov
25 April 1903 20 October 1987
A 20th-century Russian mathematician
who made significant contributions to the
mathematics of probability theory,
topology, intuitionistic logic, turbulence,
classical
mechanics,
algorithmic
information theory and computational
complexity.
62
Example
A company recruits 10 students for summer internship from twoyear full-time BM programme of XLRI for four functional areas,
namely, Analytics, Finance, Marketing and Operations. The
students are nearly equally efficient in terms of their expertise in
each of these four functional areas. On the first day of their
reporting, The HR manager in a hurry, allotted them almost at
random to the four functional areas, without taking much care
about the requirement of various areas. What is the probability
that Analytics area will receive exactly 4 of the students?
Hint Answer
A company recruits 10 students for summer internship from twoyear full-time BM programme of XLRI for four functional areas,
namely, Analytics, Finance, Marketing and Operations. The
students are nearly equally efficient in terms of their expertise in
each of these four functional areas. On the first day of their
reporting, The HR manager in a hurry, allotted them almost at
random to the four functional areas, without taking much care
about the requirement of various areas. What is the probability
that Analytics area will receive exactly 4 of the students?
Check:
Example
A large shopping complex has 15 entry gates. Usually, one security
personnel is deployed to each of these gates. Security personnel can
usually chat with their colleagues deployed in the adjacent (both right
and left) gates. The personnel in 1st and 15th gates will be able talk
with only one of their colleagues. It was observed from past CCTV
footages that two personnel, say, and whenever deployed in
adjacent gates gossips more and do not take the job seriously!
Management has ordered the chief-security officer that and
should be so deployed that there should be 10 other personnel in
between them. On one day, the chief-security officer was absent and
another person who had no idea about the order, allotted 15 personnel
in 15 gates at random. What is the probability that the requirement
will be met even in that case?
Hint Answer
Required Probability =
Example
Over the years, the culture of binge drinking spread in premier
the B-schools across the globe despite honest effort of various
managements to curb irresponsible drinking behaviour of
students. After a booze night in a hostel, it was reported that
80% of the student of one Hostel consumed Beer, 70% enjoyed
Whisky and 60% relished Vodka. What is the proportion of
stalwarts who tried all three in that night?
Poincares Theorem
For any sequence of events , (not
necessarily mutually exclusive )
Conditional Probability
may
found among the NB occurrences of B. Thus the ratio
76
provided .
If
The above definition satisfies all probability axioms discussed
earlier [Please Check by yourself]
77
(iii) Suppose are mutually disjoint for all . Then for any
Therefore
Theory of Probability
Session-4
82
83
Law of independence
86
Then and
So,
87
88
94
and
95
Bayes' Theorem
96
97
98
99