The Bernoulli Random Variable and The Binomial Distribution

Discrete Random Variable
The Bernoulli Random Variable and the Binomial Distribution
Introduction
In many situations, there are only two outcomes to a trial or experiment, one
with probability and the other with probability 1 – p. Often one outcome is called
success and other a failure. When this is the case, p is used as the probability of
success. For convenience, the probability of failure 1- p is sometimes called q.
Such an experiment is called Bernoulli trial, after Swiss mathematician James

Bernoulli or a Binomial trial , because there are only two outcomes.
The random variables associated with a Bernoulli tria is the Bernoulli random
variable, with value for a successful outcome and 0 value for failure.
Examples of Bernoulli Trials

1. Flipping a coin gives two outcomes, head and tail.
If we assign the value 1 to head and 0 to tail, we have a Bernoulli
random variable. If we call this random variable X, then P{X=1}= 0.5
and P{X=0}= 0.5
2. Rolling a die has six outcomes, but if we decide to call the event less than
or equal to two as success and call other event be greater than 3. This is
Bernoulli trial. Let Y be the random variable. Then P{ Y=1)= and P{Y=0} =
If we repeat a Bernoulli trial many time over, we get a new situation, called a
binomial experiment. The number of trials n and the probability p are its
parameters, numbers that define the structure or properties of the experiment . If
there are n rials and each trial has probability of success equal to p and failure is
equal to 1 –p, the binomial experiment denoted B(n,p). Counting the number
success in a binomial experiment gives us a new random variable, the binomial
random variable with parameters (n, p).
To compute the probability of k success in n trials, first count up how many ways
you can choose the k success and the n-k failures. This is just the number of ways to
choose k objects from a collection of n objects, nCk. Each of these choices of k objects
give an event with k success and n- k failures. Because success has probability p and
there are k success, the probability of k success is p k
Similarly, the probability of the n – k failures for each of these choices is ( 1-P) (n-
k)
.Thus for the binomial random variable X that gives the number of success for n
trials
P{ X = k}= nCk pk(1-p)(n-k)
Example. What is the probability of getting exactly 10 heads when you flip a coin 20
times? Of getting exactly eight heads?
Solution
Let getting a heads be success, the probability of heads is p= 0.5
P{ X= 10)= nCk pk(1-p)(n-k)
=20C10 (0.5)10(0.5)10
= 0.176
P{ X= 8)= nCk pk(1-p)(n-k)

=20C8 (0.5)8(0.5)12
= 0.120
We can use the binomial distribution only when there are repeated, independent
trial, each having probability of success.
The Poisson Distribution
The Poison random variable is used when the following criteria are met.
1. There is a relatively small number of success or occurences in a large interval
of time or in a large region of space.
2. The occurrences in interval or regions are independent of each other as long
as the interval or regions do not overlap.
3. The average number of occurrences in interval or region of the same size is
about the same throughout the experiment or observation.
4. Two occurrences do not happen simultaneously.
The Poison distribution is defined as
P{X = k} =
Where X is non-negative integers
λ the average rate of the outcomes that we are counting
Formula above depends on the average rate of occurrence, given as outcomes
or success per time or region. This rate is usually small and may not be an
integers.
It is sometime called the law of small numbers because it is used for
relatively rare occurrences.
Example
Left-handed people make up about 10% of the population. A school is buying
an arm-desk for a new classroom. The classroom has 30 desk, and the school wants
to put two left-handed desk and 28 right handed desks in the classroom. What is the
probability that here will be more than two left-handed students in the next class of
30?
Solution:
λ= 0.10
t = 30
λt=3
The probability that k students are left handed is
P{X = k} =
Find the complement of the event that 0, 1, 2 students will be left handed. This is
P{X = 0} = = 0.0498
P{X = 1} = = 0.149
P{X = 2} = = 0.224
P{X> 2} = 1- (0.0498 + 0.149 + 0.224)
= 0.577
What does it mean?

This is fairly large probability, so the school should consider buying more
than two left –handed desks for the classroom.
The Geometric Distribution

Suppose we flip a coin until we get a head for the first time. How many flips
will it take? With the probability of 0.5 , we will get a head on the first flipf. With the
probability of 0.5 x 0.5 = 0.25 , we will get tails then head, thus using the first head
on the second flip. The geometric distribution describes this kind of situation, when
we repeat binomial trials until we get a success.
If the probability of success in a binomial trial is p, then the geometric
random variable gives the number of trials until the first success. The probability of
n-1 failures followed by a success on the nth trial is P{ X= n}. A geometric random
variable is sometimes called a Pascal random variable.
For example, a flip a coin until you get a head. The probability of success of
getting head is 0.5. Let the random variable X give the number of trials it takes to get
the first success.
Then
X{H}=1 and P{X=1}=0.5
X{TH}=2 and P{X=2}=0.25
X{TTH}=3 and P{X=3}=0.125
The probability of getting the first head on the fifth flio is (0.5) 4 x 0.5
because the probability of failure is (0.5) 4 , and probability of success on the
last flip is 0.5. In symbol this is
P{X =5}= (0.5)4 x (0.5)
So in general the probability of the first success in nth trial is
P{X=n} = ( 1 – p)(n-1) x p
Example
Roll a die until you get a 6. What is the probability that this will
happen on the sixth roll? What is the probability that it will happen before
the sixth roll?
Solution:
P= =0.1667probability of getting a 6
1-p= 1- =0.8333probability of getting not a 6

P{X=6} = ( 0.8333)5 x 0.1667
= 0.6698
Probability that it will happen before sixth roll is
P{X=1} + P{X=2} +P{X=3} +P{X=4}+ P{X=5}
=0.1667+((0.8333)1(0.1667)(+0.8333)2(0.1667)+(0.8333)3
(0.1667)+ ( 0.8333)4 (0.1667)
= 0.5982
The Negative Binomial Distribution

We use the geometric distribution if we repeat binomial until we get the first
success but if we repeat the binomial trial until we get two the success ? or nth
success? This is the negative binomial distribution.
The Negative binomial distribution is defined as
P{X=n} = n-1C r-1 (pr ) (1-p) ( n-r)
Example:
Flip a coin until you get five heads. What is the probability of getting
five heads in 10 flips.
Solution:
In the 10 flips of course one of the success will happen so there 9C4
ways to choose four successes in the first nine flips. The probability of
success p= 0.5 and failure 1 – p= 0.5. then the probability of getting five
success in 10 flips is:
126 X ( 0,5)4 x (0.5)4 x (0.5)
The Hypergeometric Distribution

This gives the number of success when making n choices from a
collection of N objects, with probability of success equal to p. There are Np objects
that are successes, and if you want k of those , there are NpCk ways to get them.
There are N = Np other objects, and N=Np CN= k ways to choose those. There are NCn
ways to choose n objects from N objects, and this is the number of outcomes in the
sample space. This is often referred to sampling without replacement.
P={X=k} =
Example.
Suppose you choose four marbles from urn containing 20 marbles, six white
and 14 red. What is the probability to get three white and one red
Solution: The number of success is the number of ways of choosing three
white time the number of ways of choosing white marbles and that is :
6C3 x 14 C1
and the number of possible outcomes is the number of ways of choosing four
marbles from twenty that is:
20 C4
Therefore the probability to get three white and and one red is
The Multinomial Distribution

For example, if we thrown a die 10 times, what is the probability of getting a
2 three times, a 3 three times, and a 6 four times? This is an example of multinomial
distribution which is defined as :
P(n1) P(n2) P(n3). . . P(nr)

which is based on the partition rule, where P(n 1) P(n2) P(n3). . . P(nr)are the
probability for each outcomes.
Solving the problem above this will be:

= 4,200 and there are 610 outcomes so the probability is
Another example:
Marlee choose 10 candies from a bag of 100 candies. There are 10 red,
20 yellows, 25 orange, and 25 pink candies. What is the probability that she gets 3
red, 3 yellow, and 4 orange candies?
Solution:
By partition rule the number of possible partition and the
different probabilities for each outcomes are 0.10 for red ,0.20 for
yellow, 0.20 for green 0.25 for orange and 0.25 for pink
Thus the probability is
(0.10)3 (0.20)3 (0.25)4
Republic of the Philippines
BATANGAS STATE UNIVERSITY
Pablo Borbon Main I, Batangas City, Philippines 4200
COLLEGE OF TEACHER EDUCATION
Tel. No. (043) 980-0385 local 1128
Activity( Week 1)
MATH 506– Theory of Probability
Second Semester, AY 2019 – 2020
Name:__________________________________________Score: ___________
Section: _________________________________________Date: ___________
General Directions:
1. Erasures should be avoided.
2. Use black or blue pen in answering this activity. Pencil is not allowed.
3. Maintain a margin of 1inch around with 1.5” left margin.
Solve the following problems.

1. Throw a die and let 1 be a success and let 2, 3, 4, 5 and 6 be failures. This is a
Bernoulli trial.
a. What is the probability of success?
b. What is the probability of failure?
c. What does B( 6, 0.1667) mean in this context?
d. For X associated with B( 6, 0.1667) , compute P{ X=4}
e. For Y associated with B( 4, 0.1667), compute P{ Y= 1}
2. A message 10 bits long is sent. The probability of error i2 0.002 for each bit.
What is the probability that the message will be sent at most one error?
3. A random variable Y has Poison distribution P{Y=k}= e-0.3 . Find P={Y= 3}
4. It is known that there are 1,000 fish in a pond and 20 have been tagged. You
have caught three fish. What is the probability that at least one of them is
tagged?
5. A country has an average rate of three thunderstorms per week during the
summer. What is the probability that there will be at least three
thunderstorm during the next two weeks?
6. Flip a coin until you get a head. What is the probability that this will happen
on the first flip? Second flip? Third flip?
7. Toss two dice until you get 3 or 6. What is the probability that this will
happen on the first toss? Second toss? Third toss?
8. Flip a coin until you get seven heads. What is the probability that this will
happen with 10 flips? Fifteen flips” Twenty flips?
9. Toss two dice until you get a three or 6 four times. What is the probability
that this will happen on four tosses? Eight tosses? Ten tosses?
10. A bag of candies has 15 red, five yellow, 10 blue, and 20 green. Choose five
candies at random. What is the probability that you will get one red, one
yellow, one blue, and two green?
11. Toss a die six times. What is the probability of getting three 2s and three 3s?
12. For each of the following situation find at least one quantity that you could
measure that would be a random variable. For most of these, there are many
different possibilities.
a. Roll two dice
b. Choose a student at random from a class

c. Take five marbles from an urn that has red, blue, white, and green
marbles
d. Take a sample of water from a pond
e. Take a sample of earth from a garden
f. Look at the purchase at a department store on a day chosen at

random
g. Choose a batch of 10 keyboards from an assembly line that

produces hundreds of keyboards a day
h. Choose a family at random from a city.
Prepared by.
Charity Arellano Aldover Ph. D
Math 506 Assoc. Prof.

The Bernoulli Random Variable and The Binomial Distribution

Uploaded by

Copyright:

Available Formats

The Bernoulli Random Variable and The Binomial Distribution

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Bernoulli Random Variable and The Binomial Distribution

Uploaded by

Copyright:

Available Formats

Discrete Random Variable

The Bernoulli Random Variable and the Binomial Distribution

Such an experiment is called Bernoulli trial, after Swiss mathematician James

Examples of Bernoulli Trials

P{ X= 8)= nCk pk(1-p)(n-k)

The Poisson Distribution

The Poison distribution is defined as

What does it mean?

The Geometric Distribution

1-p= 1- =0.8333probability of getting not a 6

The Negative Binomial Distribution

The Hypergeometric Distribution

The Multinomial Distribution

P(n1) P(n2) P(n3). . . P(nr)

Solving the problem above this will be:

Solve the following problems.

b. What is the probability of failure?

c. What does B( 6, 0.1667) mean in this context?

d. For X associated with B( 6, 0.1667) , compute P{ X=4}

e. For Y associated with B( 4, 0.1667), compute P{ Y= 1}

3. A random variable Y has Poison distribution P{Y=k}= e-0.3 . Find P={Y= 3}

a. Roll two dice

b. Choose a student at random from a class

d. Take a sample of water from a pond

e. Take a sample of earth from a garden

f. Look at the purchase at a department store on a day chosen at

g. Choose a batch of 10 keyboards from an assembly line that

h. Choose a family at random from a city.

You might also like