0% found this document useful (0 votes)
17 views23 pages

Lecture 11.8 - Poisson Distribution

The document provides an overview of the Poisson distribution, including its probability mass function, expectation, variance, and applications in modeling events over time. It outlines learning objectives such as deriving the probability mass function and understanding situations suitable for Poisson modeling. The document also includes examples and graphical representations of the distribution's behavior based on different values of the parameter λ.

Uploaded by

bat1batttt4
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
17 views23 pages

Lecture 11.8 - Poisson Distribution

The document provides an overview of the Poisson distribution, including its probability mass function, expectation, variance, and applications in modeling events over time. It outlines learning objectives such as deriving the probability mass function and understanding situations suitable for Poisson modeling. The document also includes examples and graphical representations of the distribution's behavior based on different values of the parameter λ.

Uploaded by

bat1batttt4
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 23

Statistics for Data Science -1

Statistics for Data Science -1


Lecture: Poisson Distribution

Usha Mohan

Indian Institute of Technology Madras

1/ 27
Statistics for Data Science -1

Introduction to Poisson distribution

Probability mass function of Poisson

Expectation and variance of Poisson distribution

Applications of Poisson distribution


Modeling in time
Modeling in time

2/ 27
Statistics for Data Science -1

Learning objectives

3/ 27
Statistics for Data Science -1

Learning objectives

1. Derive the formula for the probability mass function for


Poisson distribution.

3/ 27
Statistics for Data Science -1

Learning objectives

1. Derive the formula for the probability mass function for


Poisson distribution.

3/ 27
Statistics for Data Science -1

Learning objectives

1. Derive the formula for the probability mass function for


Poisson distribution.
2. Expectation and variance of the Poisson distribution.

3/ 27
Statistics for Data Science -1

Learning objectives

1. Derive the formula for the probability mass function for


Poisson distribution.
2. Expectation and variance of the Poisson distribution.
3. To understand situations that can be modeled as a Poisson
distribution.

3/ 27
Statistics for Data Science -1
Introduction to Poisson distribution

Introduction
I The Poisson probability distribution gives the probability of a
number of events occurring in a fixed interval of time or
space.
I We assume that these events happen with a known average
rate, λ, and independently of the time since the last event.
I Let X denote the number of times an event occurs in an
interval of time (or space).
I We say X ∼ Poisson(λ), in other words, X is a random
variable that follows Poisson distribution with parameter λ.
I The Poisson distribution may be used to approximate the
Binomial distribution if the probability of success is “small”
and the number of trials is “large” .

4/ 27
Statistics for Data Science -1
Introduction to Poisson distribution

Motivation example
Consider a researcher who is observing the number of vehicles that
pass a busy traffic intersection in a day. She collects data
comprising of 1000 one minute intervals and tabulates the same in
form of a frequency table given below.
Number of vehicles 0 1 2 3 4 5 >6
Count 80 225 260 225 110 70 30

0.260
0.225 0.225

0.110
0.085
0.070

0.03

0 1 2 3 4 5 6
5/ 27
Statistics for Data Science -1
Introduction to Poisson distribution

Tabular summary

x Freq f Rel Freq fr fr x fr x 2


0 80 0.08 0 0
1 225 0.225 0.225 0.225
2 260 0.26 0.52 1.04
3 225 0.225 0.675 2.025
4 110 0.11 0.44 1.76
5 70 0.07 0.35 1.75
6 30 0.03 0.18 1.08
1000 1 2.39 7.88

I Mean= 2.39
I Variance = 7.88 − 2.392 = 2.16

6/ 27
Statistics for Data Science -1
Introduction to Poisson distribution

Observations

I Number of vehicles passing a traffic intersection are at


random and independently of each other
I The average number of vehicles per minute is about 2.39
which is equivalent to 143 per hour.
I Question: What is the appropriate probability distribution to
model the number of vehicles passing a traffic intersection?
I Poisson

7/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Derivation

Let X denote the number of events in a given interval (time or


space). Then X follows a Poisson distribution with parameter λ
1. The number of events occurring in non-overlapping intervals
are independent.
2. The probability of exactly one event in a short interval of
length, δt, is equal to λδt.
3. The probability of exactly two or more events in a short
interval is essentially zero.
What is the Probability of n events happening in interval of length
t?

8/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Poisson as Binomial approximation

I Define “success” as exactly one event happening in a short


interval of length δt
I The n events happening in interval of length t can be viewed
as n successes happening in n intervals of length δt, with each
one of them being an independent and identical trial.
 
λ
I Hence the problem can be viewed as a Bin n, p =
n
experiment.

9/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Derivation- contd

 x 
λ n−x

λ
n

= x 1−
n n
λ −x
 x 
λ n
  
n(n − 1) . . . (n − x + 1) λ
= 1− 1−
x! n n n
λ −x
x
 x  n  
λ n(n − 1) . . . (n − x + 1) λ
= 1− 1−
x! n n n
1 (x − 1)
 
x
λx  n (1 − n ) . . . (1 − )  λ n λ −x
  
= n  1−

1−
x! nx n n

Now let’s make the intervals very small, i.e, δt −→ 0 or n −→ ∞


10/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Derivation- contd

   x 
λ n−x

n λ
= lim 1−
n→∞ x n n
λx  nx (1 − n1 ) . . . (1 − (x −n 1) )  λ n λ −x
     
= lim   1− 1−
n−→∞ x! nx n n
 

| {z } | {z } | {z }
→ 1 as n → ∞ → e −λ as n → ∞ → 1 as n → ∞
λx
= e −λ
x!

11/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Probability mass function of Poisson

The distribution, with an average number of λ events per interval,


is defined as Poisson discrete random variable, X ∼ Poisson(λ),
with the p.m.f given by

e −λ λx
P(X = x) = , x = 0, 1, 2, . . .
x!

I X represents the random variable number of events per time


interval (In the example: number of vehicles passing per
minute)
I e is the mathematical constant 2.718

12/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Going back to the example

x freq Prob
0 80 0.092
1 225 0.219
2 260 0.262
3 225 0.208
4 110 0.125
5 70 0.060
6 30 0.024

13/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Observed
0.260
0.225 0.225

0.110
0.085
0.070
0.03

0 1 2 3 4 5 6

Poisson
0.262
0.219 0.208

0.125
0.092
0.0670
0.024

0 1 2 3 4 5 6
14/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Shape of pmf versus λ

I The shape of the Poisson distribution depends on the value of


the parameter λ.
I If λ is small the distribution has positive skew, but as λ
increases the distribution becomes progressively more
symmetrical.

15/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Graph of pmf for λ = 1


0.4

0.3

0.2

0.1

0
0 5 10 15
16/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Graph of pmf for λ = 4

0.2

0.15

0.1

5 · 10−2

0
0 5 10 15

17/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Graph of pmf for λ = 8


0.15

0.1

5 · 10−2

0
0 5 10 15
18/ 27
Statistics for Data Science -1
Probability mass function of Poisson

Section summary

I pmf of Poisson distribution


I shape of pmf versus λ

19/ 27

You might also like