Probability Concepts
Basic Statement on
Probability
Probability is part of our daily lives. Consider some
of the following statements
- The price of equity shares of company X is
expected to increase
- It is likely to rain heavily
Probability is the chance that a particular
event will occur.
Basic Terminology in
Probability
• Experiment: An activity that results in one and
only one outcome out of a set of disjoint
outcomes, where an outcome cannot be predicted
with certainty
• Events: The possible outcomes of an experiment
are known as events
• Sample Space: It is the set of all possible
outcomes of an experiment
Relative Frequency
Approach
– Based on statistical data
– Here, probability is defined as the
proportion of times an event occurs in
the long run when the conditions are
stable
– Alternatively, probability is defined as
the observed relative frequency of an
event in a very large number of events
Classical Approach to
Probability
• Probability can be mathematically
represented as
Number of times event occurs
P(event )
Total number of occrances
• Probability for an event A can be
P ( A) the probability
represent as of event A happening
Example
Demand for sandwiches at a Fast Food
Corner has always been either 50,100,
150, 200, or 250 per day. Over the past
200 days, the frequencies of demand are
represented in the following
Demand No. oftable:
days
50 40
100 80
150 50
200 20
250 10
Total 200
Example
Demand Frequency Probability
50 40 =40/200 = 0.2
100 80 =80/200 = 0.4
150 50 =50/200 = 0.25
200 20 =20/200 = 0.1
250 10 =10/200 = 0.05
Total Probability 1
Probability Axioms
• Each event must have associated with it a
probability greater than or equal to zero but
less than or equal to 1
0 P( A) 1
• The probability of entire sample is 1
P( S ) 1
• The probability of an event that does not
occur is equal to 1 minus probability of the
event that occurs
P( A) 1 P( A)
Types of Events
• Mutually exclusive
Events are said to be mutually exclusive if only one of
the events can occur on any one trial.
Example: a fair coin toss results in either a heads or a tails.
• Collectively Exhaustive
Events are said to be collectively exhaustive if their
union cover all the events within the entire sample
space
Example: a collectively exhaustive list of possible outcomes for a
fair coin toss includes heads and tails.
Law of Addition
Mutually Exclusive Events
P (A or B) = P (A) + P (B)
Example:
P (cricket or hockey) = P (hockey) + P (cricket)
50 50
Cricket Hockey
100
Law of Addition
Not Mutually Exclusive
Events
P(A or B) = P(A)+P(B) - P(A and B)
Example:
P (cricket or hockey) = P (hockey) + P (cricket) – P
(cricket and hockey)
40 40
C&H
Cricket 10 Hockey
100
Example
Specialized University offers four different graduate
degrees: business, education, accounting, and science.
Enrollment figures show 25 of their graduate students
are in each specialty. Although 50 of the students are
female, only 15 are female business majors. If a student
is randomly selected from the University’s registration
database:
What is the probability the student is a business or education major?
What is the probability the student is a female or a business major?
Example
The probability that the student is a business or education
major is mutually exclusive event. Thus:
P(Bus or Edu) = P(Bus) + P(Edu)
= .25 + .25
= .50
The probability that the student is a female or a business
major is not mutually exclusive because the student could
be a female business major. Thus:
P(Fem or Bus) = P(Fem) + P(Bus) – P(Fem and Bus)
= .50 + .25 - .15 = .60
Dependency
Events are either
statistically independent (the
occurrence of one event has no effect on
the probability of occurrence of the other),
or
statistically dependent (the occurrence
of one event gives information about the
occurrence of the other).
Probabilities :
Independent Events
• Marginal probability: the probability of an
event occurring: P(A)
• Joint probability: the probability of multiple,
independent events, occurring at the same time:
P(AB) = P(A)*P(B)
Probabilities :
Independent Events
• Conditional probability (for independent
events):
• the probability of event B given that event A
has occurred:
P(B|A) = P(B)
• or, the probability of event A given that event
B has occurred:
P(A|B) = P(A)
Independent Event
Example
A bag contains
3 black balls P(black ball drawn on first draw)
and 7 green P(B) =
balls. We (marginal probability)
draw a ball
P(two green balls drawn)
from the
P(GG) =
bucket,
replace it, and (joint probability for two
draw a second independent events)
ball.
Independent Event
Example
A bag contains
3 black balls
P(black ball drawn on second
and 7 green draw, first draw was green)
balls. We P(B|G) =
draw a ball (conditional probability)
from the
bucket, P(green ball drawn on second
replace it, and draw, first draw was green)
draw a second P(G|G) =
ball. (conditional probability)
Probabilities :
Dependent Events
• Marginal probability: probability of an event occurring:
P(A)
• Conditional probability :
The probability of event B given that event A has occurred:
P(B|A) = P(AB)/P(A)
The probability of event A given that event B has occurred:
P(A|B) = P(AB)/P(B)
• Joint probability: The probability of
multiple events occurring at the same
time: P(AB) = P(B|A)*P(A)
Dependent Event
Example
Then:
Assume that we
have an bag P(WL) =
containing 10 balls of
the following P(WN) =
descriptions:
4 are white (W) and P(W) =
lettered (L)
2 are white (W) and P(YL) =
numbered (N)
3 are yellow (Y) and
lettered (L)
P(YN) =
1 is yellow (Y) and
numbered (N) P(Y) =
Dependent Event
Example
Then:
We have P(Y) = 0.4
- marginal probability
P(WL) = 0.4
P(WN) = 0.2 P(L|Y) = P(YL)/P(Y)
= 0.3/0.4 = 0.75
P(W) = 0.6 - conditional probability
P(YL) = 0.3 P(W|L) = P(WL)/P(L)
P(YN) = 0.1 = 0.4/0.7 = 0.57 - -
conditional probability
P(Y) = 0.4
Combination
• A Combination is a way of selecting
several things out of a group
• It is given by following formula
n!
n
Cr
r !(n r )!
Ex: How many combinations of 5 are
possible in a standard deck of cards.
Permutation
• A permutation is an ordered combination
• It is calculated by the following formula
n!
n
Pr
(n r )!
Ex: How many colour codes can be made
out of 4 flags of different colours.
Bayes’ Theorem
• We can revise prior estimates of
probability using Bayes’ Theorem
• Ex: Suppose there are two boxes B1
and B2. B1 contains 4 white balls and
1 black ball while B2 contains 1 white
ball and 4 black balls.
Bayes’ Theorem
Let there be an event A, which can
happen, only if one of the n mutually
Bexclusive
1 , B , B .......B
2 3, n
events
occur
P ( Bi ) P( A / Bi )
P( Bi / A)
P( Bi ) P( A / Bi )
Bayes’ Theorem
Example
Three machines producing 40%, 35%
and 25% of the total output are known
to produce with defective proportion of
items as: 0.04, 0.06 and 0.03,
respectively. On a particular day, a unit
of output is selected at random, and is
found to be defective. What is the
probability that it was produced by the
second machine.
Bayes’ Theorem
Example
In a basin area where oil is likely to be found
underneath the surface, there are three locations
with three different types of earth compositions,
say C1, C2 and C3. The probabilities for these
three compositions are 0.5,0.3 and 0.2
respectively. Further it has been found from the
past experience that after drilling of well at these
locations, the probabilities of finding oil is 0.2, 0.4
and 0.3 respectively. Suppose, a well is drilled at a
location, and it yields oil, what is the probability
that the earth composition was C1.
Chebychev’s Lemma
• As per Chebychev’s Lemma, SD helps to
find out the percentage of population lying
within limits from the mean of the variable
• For any set of data, and for any constant
k>1, at least (1-1/k2) of the observations
must lie within k [Link] on either side of the
mean
Chebychev’s Lemma
• The probability that a random variable will lie
with in k [Link] of its mean is at least (1-1/k 2)
1
P ( x k x x k ) 1 2
• Ex: Let the average number of marksk
obtained by students in an examination be
60% and s.d. of their marks be 5%. Then find
what percentage of students have scores
between
mean 2 s.d .
Probability/Statisti
cal Distributions
Random Variables
• A variable is a quantity which changes or
varies due to any factor
• A variable is called as random variable
when there is a chance factor associated
with it
• The set of all possible values of random
variable and their associate probabilities
is called as probability distribution
Probability
Distribution
Ex: Let x be the number of breakdowns in machinery
of a factory in a given week. The probability
distribution for x can be given as follows
No. of P(x)
breakdowns (x)
0 0.12
1 0.2
2 0.25
3 0.3
4 0.13
Expected Value of
Random Variable
• The expected value of a random
variable is the weighted average of
all possible values that this random
variable can take on
• The weights used in computing this
average correspond to the
E ( x) x P ( x)
probabilities
Example of Expected
Value
An accountant of a company is hoping to
receive payment from two outstanding
accounts during the current month. He
estimates that there is 0.6 probability of
receiving Rs.15,000/- due from A and
0.75 probability of receiving Rs.40,000/-
due from B. What is the expected cash
flow from these two accounts.
Binomial Distribution
Visualise a practical situation where a trial
results in only two outcomes, say “Success”
and “Failure”. Further the result of one trial
does not influence the result of next trial, and
the probability of success at each trail is same.
Tossing a coin - Head or Tail
Inspection of an item - Defective or Non-defective
Repayment by a borrower - Regular or not
Binomial Distribution
The conditions for the applicability of
Binomial Distribution are
(a) There are n independent trials
(b) Each trial has only two possible outcomes
(c) The probabilities of two outcomes remain
constant
Binomial Distribution
If x is the random variable
representing number of
successes, the probability of
getting r successes and (n-r)
failures, in n trials, n is given
r n r by the
P ( x function
probability r ) Cr p q
Mean np
Std .Dev npq
Binomial Distribution
Ex: Spectrum light bulb company went on for mass
production of colour bulbs before Diwali. Five bulbs,
one of each colour, were packed in each of the boxes
which were to be sold as a unit. Due to shortage of
time, quality was compromised, and it was estimated
that 20% of the bulbs were defective.
If a customer purchases such a box of bulbs, what is
the probability that the box will have
(a) No defective bulb
(b) 2 defective bulbs
(c) At least one defective bulb
(d) At most one defective bulb
(e) All defective bulbs
Using Binomial Tables
Ex: Seven coins are tossed
simultaneously. Using binomial
distribution find out the probability of
obtaining at least five heads.
Poisson Distribution
• Distribution of rare events
• The event whose probability of
occurrence is very small but the
number of trials which could lead to the
occurrence of the event are very large
• So for a large n but very small p such
that their product is m=np, the Poisson
distribution is defined
x as
m
me
P( x)
x! e 2.718
Poisson Distribution
Ex: Suppose, we have a production
process of some item that is
manufactured in large quantities. We
find that, in general, the proportion of
defective items is p=0.01. A random
sample of 100 items is selected. What
is the probability that there are 2
defective items in this sample.
Poisson Distribution
• Ex: Suppose the probability of dialing
a wrong number is 0.05. Then, what
is the probability of dialing exactly 3
wrong numbers in 100 dials.
Normal Distribution
• Distribution for continuous random
variable
• Many real life situations can be applicable
to Normal Distribution
• Characteristics of Normal Distribution are
– Mean is located at the centre of the
distribution
– Curve is symmetrical around mean
– Mean = Median = Mode
Normal Distribution
• The probability density function of a
random variable X with
1/ 2( x m ) 2
1
P( x) e 2
2
m
Standard Normal
Distribution
• The units for standard normal
distribution curve are denoted by z
and are called as z values or z score.
• Z gives distance from mean in terms
of SD
xm
z
Properties of Normal
Distribution
Normal Distribution
Example
• Following is the data for life of bulbs
manufactured by a company.
– Mean = 1600 hrs
– S.D. = 30 hrs
(a)What is the probability that the life of a bulb
selected at random will be less than or equal
to a certain value less than the mean, say
1550 hrs?
(b)What is the proportion or percentage of bulbs
having life less than or equal to 1660 hrs?
Normal Distribution
Example
(c) What is the probability that the life of a
bulb selected at random will be between
1550 and 1580 hrs ?
(d) What is the probability that the life of a
bulb selected at random will be between
1630 and 1680 hrs ?
(e) What is the probability that the life of a
bulb selected at random will be between
1550 and 1630 hrs?
Skewness
• Skewness is a measure of the
asymmetry of the probability
distribution