Scribd
Upload
24 views

Week 5 II Discrete Random Variables and Their Distributions

Uploaded by

cool cube master chirayu
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
24 views

Week 5 II Discrete Random Variables and Their Distributions

Uploaded by

cool cube master chirayu
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Week 5 Discrete Random

Variables
Definition
• Discrete Random Variable on the probability space is defined to be a
mapping such that:

or

 The image is a countable subset of


Example
• Tossing an unfair coin three times and getting tails at least twice.
(Given )
 Let then:
Probability Mass Function (PMF)
• If is a discrete random variable with range then the members of
can be listed i.e. i.e. countable set
• So, one can define the set of outcomes , for which as an event

The corresponding probabilities over the events are known as
Probability Mass Function (PMF) of
 is known as the PMF of
 PMF is a probability measure which gives us the probabilities of the possible values for a
random variable
Probability Mass Function (PMF)
• Example 1: Tossing a fair coin 3 times. Let be defined as the number of
tails observed. Find the range of , and PMF ?
Solution: Let us find the sample space
So Range of will be number of times tails can occur i.e.
To find pmf , Define events
Find the probabilities of the events :

As the events defined above are disjoint and the event has to occur
for sure hence:
Probability Mass Function (PMF)
• Example 2: Tossing an unfair coin for which until one gets the head for the
first time.
Let be the total number of coin tosses. Find PMF
Solution: Define Sample space
Define Events
Define probabilities of these events as:

In general
Are the defined events disjoint?
Does their union cover the entire sample space , how can you verify it?
Distributions on Discrete Random Variables
What can be the simplest PMF:
 PMF that is non-zero at only one point
•
• Not very informative
 What could be the next simplest one?
• PMF that is non-zero at two points
•
where
• This is known as the “Bernoulli distribution”
Bernoulli Distribution
• It models the circumstances (involving the element of uncertainty)
that have only two possible outcomes e.g. “success” and “failure”
 Cricket fan (favouring) a team in a match
 Patient experiencing the side-effects of a new-medication in clinical trials
• In other words, Bernoulli Random Variable can be associated with
an event B such that if the event B occurs, then otherwise

 Hence the Bernoulli Random Variable is also called the Indicator random
variable
Bernoulli Distribution
• The indicator random variable for an event B has Bernoulli
Distribution with parameter and can be expressed as:
Binomial Distribution
• One tosses a coin with , times
 Let a random variable provides the total number of ‘Heads’ observed for
the above event
 Then is binomial with parameter and i.e.

 The range of , and PMF of :

where
Binomial Distribution
• Medical research: Success rate of a new treatment
• Election Forecasting: Probability of a candidate winning an election
based on the number of votes cast

Resource:
What is the real life applications of Bernoulli distribution? - Quora
Binomial Distribution Using Bernoulli
Distribution
• Consider Bernoulli random variables .
• These variables can take either value
• Associate , with the event of tossing a coin ; such that when ‘Head’
occurs and when ‘Tail’ occurs
• What will be the “event” corresponding to random variable where

 What does signify ?


 Number of heads observed in the coin tosses
• What will be the distribution of ?
 Binomial Distribution
Example
• Let and be two independent random variables. Define a new random
variable as . What will be the PMF of ?
 Can one compute it without writing anything ?
 ( is a sum of independent Bernoulli RVs)
 Similarly

• will be a sum of independent Bernoulli RVs. Hence
 What if and

Let ;
Pascal (Negative Binomial) Distribution
• One tosses a coin with , until Heads are observed
 Let a random variable provides the total number of tosses for the above
event
 Then is pascal (negative binomial) with parameter and i.e.

 The range of ,
 Consider two events

You might also like