Data Analytics

(CS40003)
Lecture #5
Probability Distributions

Dr. Debasis Samanta

Associate Professor
Department of Computer Science & Engineering
 Quote of the day..

"I avoid looking forward or backward, and try

to keep looking upward."

 CHARLOTTE BRONTE, an English novelist and poet

CS 40003: Data Analytics 2

Today’s discussion…
 Probability vs. Statistics

 Concept of random variable

 Probability distribution concept

 Discrete probability distribution

 Discrete uniform probability distribution

 Binomial distribution

 Multinomial distribution

 Hypergeometric distribution

 Poisson distribution
CS 40003: Data Analytics 3
Today’s discussion
• Continuous probability distribution
 Continuous uniform probability distribution

 Normal distribution

 Standard normal distribution

 Chi-squared distribution

 Gamma distribution

 Exponential distribution

 Lognormal distribution

 Weibull distribution

CS 40003: Data Analytics 4

 Just a minute to mark your attendance

CS 40003: Data Analytics 5

Probability and Statistics
Probability is the chance of an outcome in an experiment (also called event).

Event: Tossing a fair coin

Outcome: Head, Tail

Probability deals with predicting the Statistics involves the analysis of the
likelihood of future events. frequency of past events

Example: Consider there is a drawer containing 100 socks: 30 red, 20 blue and
50 black socks.
We can use probability to answer questions about the selection of a
random sample of these socks.
 PQ1. What is the probability that we draw two blue socks or two red socks from
the drawer?
 PQ2. What is the probability that we pull out three socks or have matching pair?
 PQ3. What is the probability that we draw five socks and they are all black?

CS 40003: Data Analytics 6

Statistics
Instead, if we have no knowledge about the type of socks in the drawers, then
we enter into the realm of statistics. Statistics helps us to infer properties about
the population on the basis of the random sample.

Questions that would be statistical in nature are:

 SQ1: A random sample of 10 socks from the drawer produced one blue, four red, five
black socks. What is the total population of black, blue or red socks in the drawer?
 SQ2: We randomly sample 10 socks, and write down the number of black socks and
then return the socks to the drawer. The process is done for five times. The mean
number of socks for each of these trial is 7. What is the true number of black socks in
the drawer?
 etc.

CS 40003: Data Analytics 7

Probability vs. Statistics
In other words:
 In probability, we are given a model and asked what kind of data we are likely to see.

 In statistics, we are given data and asked what kind of model is likely to have
generated it.

Example 4.1: Measles Study

 A study on health is concerned with the incidence of childhood measles in parents of
childbearing age in a city. For each couple, we would like to know how likely, it is that
either the mother or father or both have had childhood measles.
 The current census data indicates that 20% adults between the ages 17 and 35 (regardless of
sex) have had childhood measles.
 This give us the probability that an individual in the city has had childhood measles.

CS 40003: Data Analytics 8

Defining Random Variable
Definition 4.1: Random Variable
A random variable is a rule that assigns a numerical value to an outcome of
interest.

Example
   4.2: In “measles Study”, we define a random variable as the number
of parents in a married couple who have had childhood measles.
This random variable can take values of .
Note:
 Random variable is not exactly the same as the variable defining a data.

 The probability that the random variable takes a given value can be computed
using the rules governing probability.
 For example, the probability that means either mother or father but not both has had
measles is . Symbolically, it is denoted as P(X=1) = 0.32
CS 40003: Data Analytics 9
Probability Distribution
Definition 4.2: Probability distribution
A probability distribution is a definition of probabilities of the values of
random variable.

 
Example 4.3: Given that is the probability that a person (in the ages between
17 and 35) has had childhood measles. Then the probability distribution is given
by
X Probability

?
0 0.64
1 0.32
2 0.04

CS 40003: Data Analytics 10

Probability Distribution
 In data analytics, the probability distribution is important with which
many statistics making inferences about population can be derived .
 In general, a probability distribution function takes the following form

0.64
Example: Measles Study
0 1 2
0.64 0.32 0.04 0.32
f(x)
0.04

CS 40003: Data Analytics 11

Taxonomy of Probability Distributions
Discrete probability distributions
Binomial distribution
Multinomial distribution
Poisson distribution
Hypergeometric distribution

Continuous probability distributions

Normal distribution
Standard normal distribution
Gamma distribution
Exponential distribution
Chi square distribution
Lognormal distribution
Weibull distribution

CS 40003: Data Analytics 12

Usage of Probability Distribution
 Distribution (discrete/continuous) function is widely used in simulation studies.
 A simulation study uses a computer to simulate a real phenomenon or process as
closely as possible.

 The use of simulation studies can often eliminate the need of costly experiments and
is also often used to study problems where actual experimentation is impossible.

Examples 4.4:
1) A study involving testing the effectiveness of a new drug, the number of cured
patients among all the patients who use such a drug approximately follows a
binomial distribution.

2) Operation of ticketing system in a busy public establishment (e.g., airport), the

arrival of passengers can be simulated using Poisson distribution.

CS 40003: Data Analytics 13

 Discrete Probability Distributions

CS 40003: Data Analytics 14

Binomial Distribution
In many situations, an outcome has only two outcomes: success and failure.

 
 Such outcome is called dichotomous outcome.
 An experiment when consists of repeated trials, each with dichotomous outcome is called
Bernoulli process. Each trial in it is called a Bernoulli trial.

Example 4.5: Firing bullets to hit a target.

 Suppose, in a Bernoulli process, we define a random variable X number of successes in trials.
 Such a random variable obeys the binomial probability distribution, if the experiment satisfies
the following conditions:
1) The experiment consists of n trials.
2) Each trial results in one of two mutually exclusive outcomes, one labelled a “success” and
the other a “failure”.
3) The probability of a success on a single trial is equal to . The value of remains constant
throughout the experiment.
4) The trials are independent.

CS 40003: Data Analytics 15

Defining Binomial Distribution
 Definition 4.3: Binomial distribution

The function for computing the probability for the binomial probability
distribution is given by

for x = 0, 1, 2, …., n
Here, where denotes “the number of success” and denotes the number of
success in trials.

CS 40003: Data Analytics 16

Binomial Distribution
Example
  4.6: Measles study
X = having had childhood measles a success
p = 0.2, the probability that a parent had childhood measles
n = 2, here a couple is an experiment and an individual a trial, and the
number of trials is two.

Thus,

CS 40003: Data Analytics 17

Binomial Distribution
Example 4.7: Verify with real-life experiment
Suppose, 10 pairs of random numbers are generated by a computer (Monte-Carlo method)

15 38 68 39 49 54 19 79 38 14

If the value of the digit is 0 or 1, the outcome is “had childhood measles”, otherwise,
(digits 2 to 9), the outcome is “did not”.
For example, in the first pair (i.e., 15), representing a couple and for this couple, x = 1. The
frequency distribution, for this sample is

x 0 1 2
f(x)=P(X=x) 0.7 0.3 0.0

Note: This has close similarity with binomial probability distribution!

CS 40003: Data Analytics 18

The Multinomial Distribution
The binomial experiment becomes a multinomial experiment, if we let each trial has
more than two possible outcome.

 Definition 4.4: Multinomial distribution

If a given trial can result in the k outcomes with probabilities then the
probability distribution of the random variables representing the number of
occurrences for in n independent trials is

where =

and

CS 40003: Data Analytics 19

The Hypergeometric Distribution
 • Collection of samples with two strategies
• With replacement
• Without replacement
• A necessary condition of the binomial distribution is that all trials are
independent to each other.
• When sample is collected “with replacement”, then each trial in sample collection is
independent.

Example 4.8:
Probability of observing three red cards in 5 draws from an ordinary deck of 52
playing cards.
You draw one card, note the result and then returned to the deck of cards
Reshuffled the deck well before the next drawing is made
• The hypergeometric distribution does not require independence and is based on the
sampling done without replacement.

CS 40003: Data Analytics 20

The Hypergeometric Distribution
 In general, the hypergeometric probability distribution enables us to find the

probability of selecting successes in trials from items.

Properties of Hypergeometric Distribution

• A random sample of size is selected without replacement from items.
• of the items may be classified as success and items are classified as failure.
Let denotes a hypergeometric random variable defining the number of successes.

 Definition 4.5: Hypergeometric Probability Distribution

The probability distribution of the hypergeometric random variable , the

number of successes in a random sample of size selected from items of
which are labelled success and labelled as failure is given by

CS 40003: Data Analytics 21

Multivariate Hypergeometric Distribution
The  hypergeometric distribution can be extended to treat the case where the N

items can be divided into classes with elements in the first class , … and
elements in the class. We are now interested in the probability that a random
sample of size yields elements from elements from elements from

 Definition 4.6: Multivariate Hypergeometric Distribution

If items are partitioned into classes respectively, then the probability

distribution of the random variables , representing the number of
elements selected from in a random sample of size , is

with

CS 40003: Data Analytics 22

The Poisson Distribution
There are some experiments, which involve the occurring of the number of
outcomes during a given time interval (or in a region of space).
Such a process is called Poisson process.

Example 4.9:
Number of clients visiting a ticket selling counter in a metro station.

CS 40003: Data Analytics 23

The Poisson Distribution
Properties of Poisson process
 The number of outcomes in one time interval is independent of the number that occurs
in any other disjoint interval [Poisson process has no memory]
 The probability that a single outcome will occur during a very short interval is
proportional to the length of the time interval and does not depend on the number of
outcomes occurring outside this time interval.
 The probability that more than one outcome will occur in such a short time interval is
negligible.

 Definition 4.7: Poisson distribution

The probability distribution of the Poisson random variable , representing

the number of outcomes occurring in a given time interval , is

where is the average number of outcomes per unit time and

CS 40003: Data Analytics 24

 Descriptive measures
Given
   a random variable X in an experiment, we have denoted the probability that .
For discrete events for all values of except

Properties of discrete probability distribution

1. [ is the mean ]
2. [ is the variance ]

In summation is extended for all possible discrete values of .

Note: For discrete uniform distribution, with

and

CS 40003: Data Analytics 25

Descriptive measures
1.   Binomial distribution

The binomial probability distribution is characterized with (the probability of
success) and (is the number of trials). Then

2. Hypergeometric distribution
The hypergeometric distribution function is characterized with the size of a sample ,
the number of items and labelled success. Then

CS 40003: Data Analytics 26

Descriptive measures
3.   Poisson Distribution

The Poisson distribution is characterized with where and .

CS 40003: Data Analytics 27

 Continuous Probability
Distributions

CS 40003: Data Analytics 28

Continuous Probability Distributions

f(x)

x1 x2 x3 x4
X=x
Discrete Probability distribution

f(x)

X=x
Continuous Probability Distribution

CS 40003: Data Analytics 29

Continuous Probability Distributions
 When the random variable of interest can take any value in an interval, it is
called continuous random variable.
 Every continuous random variable has an infinite, uncountable number of possible
values (i.e., any value in an interval)
 Consequently, continuous random variable differs from discrete random
variable.

CS 40003: Data Analytics 30

Properties of Probability Density Function
The
  function is a probability density function for the continuous random
variable , defined over the set of real numbers , if

1.

f(x)

a b
X=x

CS 40003: Data Analytics 31

Continuous Uniform Distribution
 One of the simplest continuous distribution in all of statistics is the
continuous uniform distribution.

 Definition 4.8: Continuous Uniform Distribution

The density function of the continuous uniform random variable on the

interval is:

CS 40003: Data Analytics 32

Continuous Uniform Distribution
 
f(x)
c

A B
X=x

Note:
a)
b) )= where both and are in the interval (A,B)

CS 40003: Data Analytics 33

Normal Distribution
 The most often used continuous probability distribution is the normal
distribution; it is also known as Gaussian distribution.
 Its graph called the normal curve is the bell-shaped curve.

 Such a curve approximately describes many phenomenon occur in nature,

industry and research.
 Physical measurement in areas such as meteorological experiments, rainfall
studies and measurement of manufacturing parts are often more than adequately
explained with normal distribution.
 A continuous random variable X having the bell-shaped distribution is called
a normal random variable.

CS 40003: Data Analytics 34

Normal Distribution
•The mathematical equation for the probability distribution of the normal variable
 
depends upon the two parameters and , its mean and standard deviation.

f(x)

 Definition 4.9: Normal distribution

The density of the normal variable with mean and variance is

where and , the  Naperian constant

CS 40003: Data Analytics 35
Normal Distribution
σ1 = σ2
σ1

σ2

µ1
µ1 µ2
µ2 µ1 = µ2
Normal curves with µ1< µ2 and σ1 = σ2 Normal curves with µ1 = µ2 and σ1< σ2

σ1

σ2

µ1 µ2
CS 40003: Data Analytics Normal curves with µ1<µ2 and σ1<σ2 36
Properties of Normal Distribution
The

  curve is symmetric about a vertical axis through the mean
 The random variable can take any value from
 The most frequently used descriptive parameter s define the curve itself.
 The mode, which is the point on the horizontal axis where the curve is a
maximum occurs at .
 The total area under the curve and above the horizontal axis is equal to .

denotes the probability of x in the interval ().

x1 x2

CS 40003: Data Analytics 37

Standard Normal Distribution
The normal distribution has computational complexity to calculate for any two , and

 
given and
 To avoid this difficulty, the concept of -transformation is followed.

[Z-transformation]

 X: Normal distribution with mean and variance .

 Z: Standard normal distribution with mean and variance = 1.

 Therefore, if f(x) assumes a value, then the corresponding value of is given by

:
=

CS 40003: Data Analytics 38

Standard Normal Distribution
 Definition 4.10: Standard normal distribution
The distribution of a normal random variable with mean and variance is called a
standard normal distribution.

0.09
0.4
0.08 σ σ=1
0.07
0.3
0.06

0.05
0.2
0.04

0.03

0.02 0.1

0.01

0.00 0.0
-5 0 5 10 15 20 25 -3 -2 -1 0 1 2 3

x=µ µ=0
f(x: µ, σ) f(z: 0, 1)

CS 40003: Data Analytics 39

Gamma Distribution
The gamma distribution derives its name from the well known gamma function in
mathematics.

 Definition 4.11: Gamma Function

for

 Integrating by parts, we can write,

Thus function is defined as a recursive function.

CS 40003: Data Analytics 40

Gamma Distribution
 When , we can write,

Further,

Note:
[An important property]

CS 40003: Data Analytics 41

Gamma Distribution
 Definition 4.12: Gamma Distribution

The continuous random variable has a gamma distribution with parameters and
such that:

where and >0

1.0
σ=1, β=1

0.8

0.6

f(x)
0.4
σ=2, β=1

0.2
σ=4, β=1

0.0
0 2 4 6 8 10 12
x

CS 40003: Data Analytics 42

Exponential Distribution
 Definition 4.13: Exponential Distribution

The continuous random variable has an exponential distribution with parameter ,

where:
where > 0

 Note:
1) The mean and variance of gamma distribution are

2) The mean and variance of exponential distribution are

CS 40003: Data Analytics 43

Chi-Squared Distribution
 Definition 4.14: Chi-squared distribution

The continuous random variable has a Chi-squared distribution with degrees of

freedom, is given by

where is a positive integer.

 • The Chi-squared distribution plays an important role in statistical inference .

• The mean and variance of Chi-squared distribution are:

and

CS 40003: Data Analytics 44

Lognormal Distribution
The lognormal distribution applies in cases where a natural log transformation
results in a normal distribution.

 Definition 4.15: Lognormal distribution

The continuous random variable has a lognormal distribution if the random

variable has a normal distribution with mean and standard deviation The
resulting density function of is:

CS 40003: Data Analytics 45

Lognormal Distribution

0.7

µ=0, σ=1
0.6

0.5

0.4
f(x)
0.3

0.2
µ=1, σ=1
0.1

0.0
0 5 10 15 20
x

CS 40003: Data Analytics 46

Weibull Distribution
 Definition 4.16: Weibull Distribution

The continuous random variable has a Weibull distribution with parameter and
such that.

where and

1.0
ß=1
The
  mean and variance of Weibull
0.8
distribution are:

0.6

f(x)
0.4
ß=2

0.2
ß=3.5

0.0
0 2 4 6 8 10 12
x

CS 40003: Data Analytics 47

 Reference

The detail material related to this lecture can be found in

Probability and Statistics for Enginneers and Scientists (8 th Ed.) by

Ronald E. Walpole, Sharon L. Myers, Keying Ye (Pearson), 2013.

CS 40003: Data Analytics 48

 Any question?

You may post your question(s) at the “Discussion Forum”

maintained in the course Web page!

CS 40003: Data Analytics 49

 Questions of the day…
1. Give some examples of random variables? Also, tell the
range of values and whether they are with continuous or
discrete values.

2. In the following cases, what are the probability

distributions are likely to be followed. In each case, you
should mention the random variable and the parameter(s)
influencing the probability distribution function.
a) In a retail source, how many counters should be opened at a given
time period.
b) Number of people who are suffering from cancers in a town?

CS 40003: Data Analytics 50

 Questions of the day…
2. In the following cases, what are the probability
distributions are likely to be followed. In each case, you
should mention the random variable and the parameter(s)
influencing the probability distribution function.
c) A missile will hit the enemy’s aircraft.
d) A student in the class will secure EX grade.
e) Salary of a person in an enterprise.
f) Accident made by cars in a city.
g) People quit education after i) primary ii) secondary and iii) higher
secondary educations.

CS 40003: Data Analytics 51

 Questions of the day…
3. How you can calculate the mean and standard deviation of
a population if the population follows the following
probability distribution functions with respect to an event.
a) Binomial distribution function.
b) Poisson’s distribution function.
c) Hypergeometric distribution function.
d) Normal distribution function.
e) Standard normal distribution function.

CS 40003: Data Analytics 52