Theory of Probability and

Mathematical Statistics

Week 4
Types of distribution for DRV

Akerke Zhailaubek
[email protected]

Lecture overview:

1. The Binomial distribution.

2. The mean and standard deviation of the
binomial distribution. The Bernoulli's
formula.
3. The Poisson probability distribution.
4. The mean and variance of Poisson
distribution.

 Real Life Applications of Binomial Distribution

§ A new drug introduced will
either cure patient or not
§ In industrial processes,
equipment breaks down or not
on a certain step
§ In communication networks, a
line is busy or not
§ In game theory, the player
either wins or not etc.

46
 Binomial Distribution
§ The binomial distribution with parameters n
and p is the discrete probability distribution of
the number of successes p in a sequence of n
independent experiments
§ Each experiment has only two possible
outcomes (true-false, yes–no, success-failure
etc)

47
 Example

In a clinical treatment trial, a patient’s

condition may improve or not. We study
the number of patients who improved,
but not how much they feel better.

48
 When a binomial distribution is a
suitable model
§ The number of trials n must be fixed. For
instance, coin was thrown 10 times.
§ Each trial must have all outcomes classified
into two categories (commonly referred to
as success and failure, yes and no, accepted
and rejected, head or tail)
§ Trials are independent (this allows us to
multiply probabilities)
§ Probability of success remains the same
for all trials
49
 Binomial Theorem
§ The binomial theorem (or binomial
expansion) describes the algebraic
expansion of powers of a binomial.
§ According to the theorem, it is possible
-
to expand the polynomial (𝑎 + 𝑏) into a
sum involving terms
- . -/.
.𝐶𝑎 𝑏 ,
where coefficients .-𝐶 are specific positive
numbers called binomial coefficients
 Binomial Coefficients
§ Binomial coefficients can be found using
Pascal’s triangle
§ Each coefficient in the
triangle is equal to the sum
of the two numbers above
(𝑎 + 𝑏)% = 𝑎 + 𝑏
(𝑎 + 𝑏)" = 𝑎" + 2𝑎𝑏 + 𝑏 "
(𝑎 + 𝑏)# = 𝑎# + 3𝑎" 𝑏 + 3𝑎𝑏 " + 𝑏 #
(𝑎 + 𝑏)$ = 𝑎$ + 4𝑎# 𝑏 + 6𝑎" 𝑏 " + 4𝑎𝑏 # + 𝑏 $
Binomial Coefficients
§ Pascal’s triangle is not convenient for
large exponents
§ Each binomial coefficient can be found
using the factorial equation
𝑛 𝑛!
= -"𝐶 =
𝑥 𝑥! 𝑛 − 𝑥 !
for 𝑥 = 0,1,2,3, … 𝑛 , where
𝑛 = number of trials
𝑥 = number of successes in 𝑛 trials
Why Binomial Coefficients
§ Why do we need binomial coefficients in
this lecture?

Example
A fair coin is tossed five times. What is the
probability to obtain exactly 3 heads in five
tosses?
 Example. Solution
§ Three heads in five trials can be obtained in
the following ten ways THHHT
HHHTT
§ This is the number of TTHHH
HTTHH
arrangements of 5 objects HHTTH
with 2 (TT) and 3 (HHH) THTHH
HTHTH
identical objects HHTHT
THHTH
§ The number of distinct HTHHT

combinations is given by the

binomial coefficient 5 5!
= 8:𝐶 = = 10
3 3! 2!
 Example. Solution
§ The probability of each of the ten outcomes
is 1 1 1 1 1 THHHT
𝑃=
2 2 2 2 2 HHHTT
TTHHH
§ Thus, the probability to obtain HTTHH
HHTTH
exactly three Heads (𝑥 = 3 𝐻) in THTHH
HTHTH
the experiment is given by HHTHT
& &
&
1 5 1 THHTH
𝑃 𝑥=3 = #𝐶 = HTHHT
2 3 2
§ This is why we need binomial coefficients
 Example. Solution
Please note that:
§ The number of trials is fixed (5)
§ There are only two possible outcomes
(Head and Tail)
§ The probability of each outcome does
not change (a fair coin for all trials)
§ The results of each trial are independent
of each other
 General Binomial Distribution
Formula
§ In general, the probability to have exactly
𝑥 successes in 𝑛 trials is given by the formula
Probability to have
Number of trials
exactly 𝑥 successes

𝑛 " -/"
𝑃 𝑋=𝑥 = 𝑝 (1 − 𝑝)
𝑥
Number of Probability of Probability
successes success of failure
 Probability Distribution Function
§ The formula in the previous slide is called
Probability Distribution Function (PDF) or
Probability Mass Function (PMF) and can
be rewritten as
𝑛!
𝑃 𝑋=𝑥 = 𝑝 " (1 − 𝑝)-/"
𝑥! 𝑛 − 𝑥 !
§ Sometimes 𝒏 is called the index and 𝒑 is
the parameter of the binomial distribution
 Example

A fair die is rolled 10 times. Find the

probability of:
a) no sixes to appear on upper face
%'
10! 5
𝑃 𝑛𝑜 6 = × = 0.1615
0! 10! 6
b) only 2 sixes to appear
" )
(
10! 1 5
𝑃 𝑡𝑤𝑜 6 𝑠 = × = 0.29
2! 8! 6 6
 Short Notation for Binomial
Distribution
§ So, the binomial distribution is described
by the lengthy formula
𝑛! " -/"
𝑃 𝑋=𝑥 = 𝑝 (1 − 𝑝)
𝑥! 𝑛 − 𝑥 !
§ Sometimes a short notation is used
𝑋~𝐵(𝑛, 𝑝)
§ This means a discrete random variable 𝑋 is
binomially distributed with number of trials
𝒏 and probability for success 𝒑
 Binomial Cumulative Distribution Function

§ Binomial cumulative distribution function

(CDF) is described by the formula
,
𝑛!
𝑃 𝑋≤𝑥 =B 𝑝 , (1 − 𝑝)-.* = 1
𝑘! 𝑛 − 𝑘 !
*+'

§ Binomial CDF can be used to calculate

probability that a discrete random variable
is less (or greater) than a certain value
Example
A random variable is described by
&
Binomial distribution 𝑋~𝐵 12, . Find
0
a) 𝑃(𝑋 = 2)
b) 𝑃(𝑋 < 1)
c) 𝑃(𝑋 ≤ 1)
d) 𝑃(𝑋 ≥ 1)
e) 𝑃(𝑋 ≤ 10)
Example
A random variable is described by
&
Binomial distribution 𝑋~𝐵 12, 0 . Find
7 ! 8 79
a) 𝑃 𝑋=2 = 7!
! ; ;
≈ 0.296
7 9 8 7!
b) 𝑃 𝑋 < 1 =𝑃 𝑋=0 = 7!
9 ; ;
≈ 0.112
c) 𝑃 𝑋 ≤ 1 =P X=0 +P X=1 =
12 1 9 5 7! 12 1 7 5 77
+ ≈ 0.381
0 6 6 1 6 6
Example
A random variable is described by
&
Binomial distribution 𝑋~𝐵 12, 0 . Find
d) 𝑃 𝑋 ≥1 =1−𝑃 𝑋 =0 =1−
7! 7 9 8 7!
9 ; ;
≈ 0.888
e) 𝑃𝑋 ≤ 10 = 1 − 𝑃 𝑋 = 11 − 𝑃 𝑋 = 12 =
12 1 77 5 7 12 1 7! 5 9
1− − ≈ 0.999 …
11 6 6 12 6 6
 The Expected Value of Binomial Distribution
§ The expected value of a binomially distributed discrete
random variable 𝑋 can be found using the definition of
the mean
-

𝐸 𝑋 = 6 𝑥𝑃(𝑥)
"1$
§ There is also a shorter way. If 𝑋 ~ 𝐵(𝑛, 𝑝), then the
expected value of 𝑋 is
𝐸 𝑋 = 𝑛𝑝
 The Variance of Binomial Distribution

§ The variance of a binomially distributed discrete

random variable 𝑋 can be found using the definition of
the variance

𝑉𝑎𝑟 𝑋 = 𝐸 𝑋 * − 𝐸(𝑋) *

§ Similarly, there is a shorter way. If 𝑋 ~ 𝐵(𝑛, 𝑝), then

the variance of 𝑋 is
𝑉𝑎𝑟 𝑋 = 𝑛𝑝(1 − 𝑝)
 Example
A fair coin is tossed three times. Find the
mean and the variance. 𝑛 = 3, 𝑝 = 0.5
# of Heads, 0 1 2 3
𝑥
𝑥! 0 1 4 9
𝑃(𝑋 = 𝑥) 1 3 3 1
$
8 8 8 8
1 3 3 1 12
𝐸 𝑋 = & 𝑥𝑃 𝑥 = 0× + 1× + 2× + 3× = = 1.5
8 8 8 8 8
!"#

𝐸 𝑋 = 𝑛𝑝 = 3×0.5 = 1.5
 Example
Find the variance.
# of Heads, 𝑥 0 1 2 3
𝑥! 0 1 4 9
𝑃(𝑋 = 𝑥) 1 3 3 1
8 8 8 8
% 1 3 3 1
𝑉𝑎𝑟 𝑋 = 𝐸 𝑋% − 𝐸 𝑋 = 0× + 1× + 4× + 9× − 1.5%
8 8 8 8
3
= = 0.75
4

𝑉𝑎𝑟 𝑋 = 𝑛𝑝(1 − 𝑝) = 3×0.5×(1 − 0.5) = 0.75

 Exercise
A fair die is tossed 𝑛 = 6 times. Probability to get 6 on upper
7
face is 𝑝 = . Find the mean and the variance.
;

# of Sixes, 0 1 2 3 4 5 6
𝑥
𝑥! 0 1 4 9 16 25 36
𝑃(𝑋 = 𝑥) ! " # $ % & ' ' & % $ # " !
1 5 1 5 1 5 1 5 1 5 1 5 1 5
1 6 15 20 15 6 1
6 6 6 6 6 6 6 6 6 6 6 6 6 6

1
𝐸 𝑋 = 𝑛𝑝 = 6× = 1
6
 Exercise. Solution
Also expected value
# of Sixes, 0 1 2 3 4 5 6
𝑥
𝑥! 0 1 4 9 16 25 36
𝑃(𝑋 = 𝑥) ! " # $ % & ' ' & % $ # " !
1 5 1 5 1 5 1 5 1 5 1 5 1 5
1 6 15 20 15 6 1
6 6 6 6 6 6 6 6 6 6 6 6 6 6

& # & ' (

1 5 1 5
𝐸 𝑋 = & 𝑥𝑃(𝑥) = 0×1 + 1×6 +
6 6 6 6
!"#
)
1 %
5 1 $ 5 $ 1 ) 5 %
2×15 + 3×20 + 4×15 +
6 6 6 6 6 6
1 (
5 ' 1 & 5 # 46656
5×6 + 6×1 = =1
6 6 6 6 46656
 Exercise. Solution
A fair die is tossed 𝑛 = 6 times. Probability to get 6 on upper
7
face is 𝑝 = . Find the mean and the variance.
;

# of Sixes, 0 1 2 3 4 5 6
𝑥
𝑥! 0 1 4 9 16 25 36
𝑃(𝑋 = 𝑥) ! " # $ % & ' ' & % $ # " !
1 5 1 5 1 5 1 5 1 5 1 5 1 5
1 6 15 20 15 6 1
6 6 6 6 6 6 6 6 6 6 6 6 6 6

1 1
𝑉𝑎𝑟 𝑋 = 𝑛𝑝(1 − 𝑝) = 6× × 1 − = 0.833
6 6
 Exercise. Solution
Also variance
# of Sixes, 0 1 2 3 4 5 6
𝑥
𝑥! 0 1 4 9 16 25 36
𝑃(𝑋 = 𝑥) ! " # $ % & ' ' & % $ # " !
1 5 1 5 1 5 1 5 1 5 1 5 1 5
1 6 15 20 15 6 1
6 6 6 6 6 6 6 6 6 6 6 6 6 6

# & ' (
% % 1 5 1 5
𝑉𝑎𝑟 𝑋 = 𝐸 𝑋 − 𝐸 𝑋 = 0×1 + 1×6 +
6 6 6 6
) $ %
1 % 5 1 $ 5 1 ) 5
4×15 + 9×20 + 16×15 +
6 6 6 6 6 6
1 ( 5 ' 1 & 5 # 85536 5
25×6 + 36×1 −1= − 1 = = 0.833
6 6 6 6 46656 6
Example
An archer arrows at a target and for each arrow,
independently of all the others, the probability of not
hitting the bull's eye is 7/8.
a) Given that the archer fires 5 arrows, find the probability
that fewer than 2 arrows hit the bull's eye.
The archer fires 5 arrows, collects them from a target and
fires all 5 again.
b) Find the probability that on both occasions fewer than
2 hit the bull's eye.
Example. Solution
a) Given that the archer fires 5 arrows, find the probability
that fewer than 2 arrows hit the bull's eye.
𝑋 is the number of hits in five shots
1
𝑋~𝐵 5,
8
𝑃 𝑋 <2 =𝑃 𝑋 ≤1 =𝑃 𝑋 =0 +𝑃 𝑋 =1
7 8 1 7 <
= + 5× = 0.87927
8 8 8
Example. Solution.
The archer fires 5 arrows, collects them from a target and
fires all 5 again.
b) Find the probability that on both occasions fewer than
2 hit the bull's eye.
In both series of shots the number of hits is
less than 2, i.e. 𝑋 < 2
𝑃 = 𝑃 𝑋 < 2 ×𝑃 𝑋 < 2 = 0.87927 ! = 0.773
Exercise
A fair 4-sided die has numbers 1, 2, 3 and 4 on its faces.
The die is rolled 20 times. The random variable
𝑋 represents the number of 4s obtained.
a) Find the mean 𝜇 and the variance of 𝑋
𝐸 𝑋 = 𝜇 = 𝑛𝑝 = 20×0.25 = 5
𝑉𝑎𝑟 𝑋 = 𝑛𝑝 1 − 𝑝 = 20×0.25×0.75 = 3.75
b) Find 𝑃(𝑋 < 𝜇 − 𝑉𝑎𝑟(𝑋))
𝑃 𝑋 < 𝜇 − 𝑉𝑎𝑟 𝑋 = 𝑃 𝑋 ≤ 3 = 0.2252
Real life applications of Poisson Distribution

Poisson distribution is applied in situations

where there is a large number of independent
Bernoulli experiments (two possible outcomes
only) with a probability ≪ 1 of success like
§ The number of aircraft/road accidents in any
time interval
§ The number of defects in an electronic
device
§ The number of persons suffering from any
rare disease in a large population
§ The number of beds occupied in a large
hospital etc
3
 Poisson Distribution

§The Poisson distribution is a discrete

probability distribution that expresses the
probability of a given number of events
occurring in a fixed interval of time or
space if these events occur with a known
constant mean rate and independently of
the time since the last event
 Number of flaws Frequency
0 12
1 29
2 25
3 17
4 11
5 5
6 1
We are going to attempt to find a probability
distribution that models this data.
5
What is the mean
and variance?

6
The mean number of flaws per sheet is
205/100 = 2.05

and the variance is

619/100 – 2.052 = 1.98

The mean and variance are both

approximately 2.
Let’s introduce some probability using ‘relative
frequency.
7
 Number of Frequency Relative
flaws frequency
0 12 0.12
1 29 0.29
2 25 0.25
3 17 0.17
4 11 0.11
5 5 0.05
6 1 0.01
Let 𝑿 be the discrete random variable,
number of flaws on a single sheet of metal.
8
Table summarising the probability distribution for
the number of observed flaws counted on each of
the 100 metal sheets

x 0 1 2 3 4 5 6
P(X=x) 0.12 0.29 0.25 0.17 0.11 0.05 0.01

9
 We want to find a probability function
𝑷(𝑿 = 𝒙) that will give us a similar table.
𝟎 ≤ 𝑷(𝑿 = 𝒙) ≤ 𝟏

∑𝒂𝒍𝒍 𝒙 𝑷(𝑿 = 𝒙) = 𝟏

Why all x and not x= 0 to x=6?

The table shows 𝒙 = 𝟔 is the largest value but in a
different sample it might be less than 6 or even
more than six. The greatest number of flaws is
unknown. 𝒙 = 𝟎, 𝟏, 𝟐, ……
10
So we want the sum of an infinite series
to equal 1 and contain only positive
terms between 0 and 1 .

&

0𝑷 𝑿 = 𝒙 = 𝟏
𝒙$𝟎

11
Lets begin with a simple mathematical relationship
𝒂𝝀 ×𝒂"𝝀 = 𝒂𝟎 = 𝟏
Where 𝒂 and 𝜆 are positive constants

We want to have an infinite series though so why not

begin with
𝒆𝝀 ×𝒆"𝝀 = 𝒆𝟎 = 𝟏
because we know how to express 𝒆𝝀 as an infinite
series.

12
 𝝀𝟎 𝝀𝟏 𝝀𝟐 𝝀𝟑 𝝀𝟒
𝑵𝒐𝒘 𝒆𝝀 = + + + + +⋯
𝟎! 𝟏! 𝟐! 𝟑! 𝟒!

𝝀𝟎 𝝀𝟏 𝝀𝟐 𝝀𝟑
𝒆-𝝀 ×𝒆𝝀 = 𝒆-𝝀 + + + +. . =𝟏
𝟎! 𝟏! 𝟐! 𝟑!
&
.
𝜆
= 0 𝑒 -0 =1
𝑥!
.$/

13
This is an infinite series whose sum is one and
contains only positive terms which lie between
0 and 1.
0 &
So 𝑃 𝑋=𝑥 = 𝑒 -0
.!
is a probability function for any positive
constant 𝜆.

The 𝒙 is the value of the random variable but

what can we say about 𝝀? Let’s find the mean
value of the discrete random variable 𝑿.
14
 $ $ $
𝝀𝒙 𝝀𝒙
𝑬 𝑿 = ' 𝒙 𝑷(𝑿 = 𝒙) = ' 𝒙 𝒆%𝝀 = ' 𝒙 𝒆%𝝀
𝒙! 𝒙!
𝒙"𝟎 𝒙"𝟎 𝒙"𝟏

$ $
𝝀𝒙%𝟏 𝝀𝒙%𝟏
𝑬(𝑿) = ' 𝒆%𝝀 𝝀 = 𝝀𝒆%𝝀 '
𝒙−𝟏 ! 𝒙−𝟏 !
𝒙"𝟏 𝒙"𝟏
𝝀𝟐 𝝀𝟑
= 𝝀𝒆%𝝀 𝟏+𝝀+ + +⋯
𝟐! 𝟑!
𝑬 𝑿 = 𝝀𝒆!𝝀 𝒆𝝀

𝑬 𝑿 =𝝀

The constant 𝝀 is the mean, which is very convenient.

Do you remember the example we started the lecture with ? What

was the mean ?
15
Here is the observed data again
X 0 1 2 3 4 5 6
P(X=x) 0.12 0.29 0.25 0.17 0.11 0.05 0.01
This is the model we are going to use
𝒙
𝝀
𝑷 𝑿 = 𝒙 = 𝒆-𝝀
𝒙!
𝒙
𝟐
𝑷 𝑿 = 𝒙 = 𝒆-𝟐
𝒙!
Now calculate and compare
16
 Observed
𝑿 0 1 2 3 4 5 6
𝑃(𝑋 = 𝑥) 0.12 0.29 0.25 0.17 0.11 0.05 0.01

Predicted using model

𝑿 0 1 2 3 4 5 6
𝑃(𝑋 = 𝑥) 0.14 0.27 0.27 0.18 0.09 0.04 0.01

17
The model matches closely with the experimental
data.

The probability function that defined the model,

𝝀 𝒙
-𝝀
𝑷 𝑿=𝒙 =𝒆 for 𝒙 = 𝟎, 𝟏, 𝟐 … … is called
𝒙!
the Poisson probability function.

The random variable 𝑿 has a Poisson distribution

with parameter 𝝀.
18
Poisson Probability Function

Wikipedia.org
We have already shown that the mean of a random
variable which has a Poisson distribution with
parameter 𝝀 is 𝝀 .

The variance is also equal to 𝝀

20
𝑽𝒂𝒓 𝑿 = 𝑬 𝑿𝟐 − (𝑬 𝑿 )𝟐

𝑬 𝑿𝟐 = ∑𝒂𝒍𝒍 𝒙 𝒙𝟐 𝑷(𝒙 = 𝒙)
𝝀 𝒙
𝑬 𝑿𝟐 = ∑&
𝒙$𝟎 𝒙 𝟐
𝒆 -𝝀
𝒙!

&
𝒙
𝝀
𝑬(𝑿𝟐 ) = 0 𝒙 𝒙 − 𝟏 + 𝒙 𝒆-𝝀
𝒙!
𝒙$𝟎

21
 &
𝒙 𝒙
𝝀 𝝀
𝑬 𝑿𝟐 = 0 𝒙 𝒙 − 𝟏 𝒆-𝝀 + 𝒙𝒆-𝝀
𝒙! 𝒙!
𝒙$𝟎

& &
𝒙 𝒙
𝝀 𝝀
= 0 𝒙 𝒙 − 𝟏 𝒆-𝝀 + 0 𝒙𝒆-𝝀
𝒙! 𝒙!
𝒙$𝟎 𝒙$𝟎

& &
𝒙 𝒙
𝝀 𝝀
= 0 𝒙 𝒙 − 𝟏 𝒆-𝝀 + 0 𝒙𝒆-𝝀
𝒙! 𝒙!
𝒙$𝟐 𝒙$𝟏

22
 $ 𝒙 $ 𝒙
𝝀 𝝀
𝑬(𝑿𝟐) = & 𝒙 𝒙 − 𝟏 𝒆%𝝀 + & 𝒙𝒆%𝝀
𝒙! 𝒙!
𝒙"𝟐 𝒙"𝟏

$ 𝒙%𝟐 $ 𝒙%𝟏
𝝀 𝝀
= & 𝝀𝟐 𝒆%𝝀 + & 𝝀𝒆%𝝀
𝒙−𝟐 ! 𝒙−𝟏 !
𝒙"𝟐 𝒙"𝟏

$ 𝒙%𝟐 $
𝒙%𝟏
𝝀 𝝀
= 𝝀𝟐 𝒆%𝝀 & + 𝝀𝒆%𝝀 &
𝒙−𝟐 ! 𝒙−𝟏 !
𝒙"𝟐 𝒙"𝟏

= 𝝀𝟐 𝒆%𝝀 𝒆𝝀 + 𝝀𝒆%𝝀 𝒆𝝀 = 𝝀𝟐 + 𝝀
23
 𝟐 𝟐
𝑽𝒂𝒓 𝑿 = 𝑬 𝑿 − 𝑬 𝑿

𝑽𝒂𝒓 𝑿 = 𝝀𝟐 + 𝝀 − 𝝀𝟐 = 𝝀

Remember this as a fact.

24
𝑻𝒉𝒆 𝒓𝒂𝒏𝒅𝒐𝒎 𝒗𝒂𝒓𝒊𝒂𝒃𝒍𝒆 𝑿 𝒉𝒂𝒔 𝒂
𝑷𝒐𝒊𝒔𝒔𝒐𝒏 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏

𝒆(𝝀 𝝀𝒙
𝑷 𝑿=𝒙 = 𝒙 = 𝟎, 𝟏, 𝟐 …
𝒙!

𝒎𝒆𝒂𝒏 = 𝒗𝒂𝒓𝒂𝒊𝒏𝒄𝒆 = 𝝀

𝑿~𝑷𝒐 (𝝀)

25
Situations that can be modelled by a
Poisson distribution

The random variable 𝑿 represents the number of

events that occur in an interval of space
(length, area, volume) or time.

Examples flaws in a length of material

calls received by a single telephone
breakdown on a particular machine
26
The events we are counting must be occurring
randomly:

Two events can’t happen at the same time.

Events occur singly in space or time.

Events are independent (no bursts or clusters).

Events occur at a constant rate e.g. If the mean

rate is 2 calls a minute the mean rate for a 10
minute interval will be 20 calls.
27
Example
A shop sells a particular product at
a rate of 2.5 per week. Find the
probability that in a two- week
period the shop sells;
(i) 3 products,
(ii) at least 3 products.

Assume that a Poisson model is

appropriate. 28
The given rate is 2.5 per week but the
question is asking about an interval of 2
weeks. The mean rate for two weeks is 5.
𝒆-𝝀 𝝀𝒙
𝑷 𝑿=𝒙 = 𝒙 = 𝟎, 𝟏, 𝟐 …
𝒙!

𝒆-𝟓 𝟓𝒙
𝑷 𝑿=𝒙 = 𝒙 = 𝟎, 𝟏, 𝟐 …
𝒙!

𝒆-𝟓 𝟓𝟑
𝑷 𝑿=𝟑 = = 𝟎. 𝟏𝟒𝟎𝟒
𝟑!
29
𝑷 𝑿 ≥ 𝟑 = 𝑷 𝑿 = 𝟑 + 𝑷 𝑿 = 𝟒 +. . . .

𝑷 𝑿≥𝟑 =𝟏−𝑷 𝑿≤𝟐

𝟓𝟐
𝑷 𝑿 ≥ 𝟑 = 𝟏 − 𝒆-𝟓 𝟏+𝟓+
𝟐!

𝑷 𝑿 ≥ 𝟑 = 𝟎. 𝟖𝟕𝟓𝟑

30
BINOMIAL POISSON
𝒙 = 𝟎, 𝟏 … 𝒏 𝒙 = 𝟎, 𝟏 … . .
n+1 terms Infinite terms

Two parameters One parameter

𝒏 and 𝒑 𝝀

mean = 𝒏𝒑 mean = 𝝀

variance = 𝒏𝒑𝒒 variance = 𝝀

32
 Decide whether binomial or Poisson is an
appropriate model
Binomial Poisson
Fixed number of trials Average rate of
occurrences in a
Constant probability of specified interval
success of time or space.

If the number of successes

is known then the number
of failures is also known
40
Example
A machine produces a continuous length of
of cloth which has on average a single defect
every 20m. Find the probability, to three
decimal places, that in a length of 10m there
are not any defects.

Cloth without defects is sold at premium

price. A length of 100m is cut into lengths of
10m. What is the probability that at least
three of these 10m lengths will be sold at
premium price.
41
The first part of the question is Poisson “average
1 per 20m”

The second part of the question is binomial

because we have 10 pieces of cloth.

42
1 defect every 20m so 0.5 defects every 10m
(average). If X is the number of defects in 10m
of material then 𝑿~𝑷𝒐 (𝟎. 𝟓)
𝒙
𝟎. 𝟓
𝑷 𝑿 = 𝒙 = 𝒆-𝟎.𝟓
𝒙!

𝑷 𝑿 = 𝟎 = 𝒆-𝟎.𝟓 = 𝟎. 𝟔𝟎𝟕

43
10 pieces of cloth and the probability that there
are not any defects on a single piece is 0.607. If X
is the number of pieces of cloth without defects
then 𝑿~𝑩 (𝟏𝟎, 𝟎. 𝟔𝟎𝟕)

𝟏𝟎!
𝑷 𝑿=𝒙 = 𝟎. 𝟔𝟎𝟕𝒙 𝟎. 𝟑𝟗𝟑𝟏𝟎-𝒙
𝒙! 𝟏𝟎-𝒙 !

𝑷 𝑿 ≥ 𝟑 = 𝟏 − 𝑷 𝑿 ≤ 𝟐 = 𝟎. 𝟗𝟖𝟗

44
Summary
𝑰𝒇 𝒂 𝒓𝒂𝒏𝒅𝒐𝒎 𝒗𝒂𝒓𝒊𝒂𝒃𝒍𝒆 𝑿 𝒉𝒂𝒔 𝒂
𝑷𝒐𝒊𝒔𝒔𝒐𝒏 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏 𝒕𝒉𝒆𝒏

𝒆(𝝀 𝝀𝒙
𝑷 𝑿=𝒙 = 𝒙 = 𝟎, 𝟏, 𝟐 …
𝒙!

𝒎𝒆𝒂𝒏 = 𝒗𝒂𝒓𝒂𝒊𝒏𝒄𝒆 = 𝝀

𝑿~𝑷𝒐 (𝝀)
45
References:
1. Palin A., Park A., Whiteley C., (2012), A-level
mathematics for Edexcel Statistics 1, CGP, UK.
2. Attwood, G., Clegg, A., Dyer, G. and Dyer, J
(2008), Edexcel AS and A-Level Modular
Mathematics series S2, Pearson, Harlow, UK.
3. Lecture notes, NUFYP, Nazarbayev University.

61