Lesson 8

THE POISSON PROBABILITY DISTRIBUTION

The Poisson Distribution was developed by the French mathematician Simeon Denis
Poisson in 1837
Poisson distribution is sometimes called a 𝑷𝒐𝒊𝒔𝒔𝒐𝒏𝒊𝒂𝒏.

The Poisson random variable satisfies the following conditions:

 The number of successes in two disjoint time intervals is independent

 The probability of a success during a small-time interval is proportional to the

entire length of the time interval apart from disjoint time intervals, the Poisson
random variable also applies to disjoint regions of space
Occurrence
Applications of the Poisson distribution can be found in many fields related to
counting
• Electrical system example: telephone calls arriving in a system.
• Astronomy example: photons arriving at a telescope.
• Biology example: the number of mutations on a strand of DNA per unit length.
• Management example: customers arriving at a counter or call centre.
• Civil Engineering example: cars arriving at a traffic light.
• Finance and Insurance example: Number of Losses/Claims occurring in a given
period of Time.
• Earthquake Seismology example: An asymptotic Poisson model of seismic risk
for large earthquakes. (Lomnitz, 1994)
Applications
 the number of deaths by horse kicking in the Prussian army (first application)
 birth defects and genetic mutations

 rare diseases (like Leukaemia, but not AIDS because it is infectious and so
not independent) - especially in legal cases
 car accidents
 traffic flow and ideal gap distance
 number of typing errors on a page
 hairs found in McDonald's hamburgers
 spread of an endangered animal in Africa
 failure of a machine in one
month
When the number 𝐧 of trials is very large and the probability 𝐩 small, e.g. 𝐧 > 25
and 𝐩 < 0.1, binomial probabilities are often approximated by the Poisson distribution

The probability distribution of a Poisson random variable X representing the number of

successes occurring in a given time interval or a specified region of space is given by
the formula:
𝑒 −𝜇 𝜇𝑥
𝑃 𝑋 = 𝑥!

Where
𝑥 = 0, 1, 2, 3, …
𝑒 = 𝐸𝑢𝑙𝑒𝑟 ′ 𝑠 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
= 2.71828 (𝑏𝑢𝑡 𝑤𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑙𝑦 𝑢𝑠𝑒 𝑜𝑢𝑟 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑜𝑟 ′ 𝑠 𝑒 𝑏𝑢𝑡𝑡𝑜𝑛)
𝜇 = mean number of successes in the given time interval or region of space
Mean and Variance of Poisson Distribution
If 𝜇 is the average number of successes occurring in a given time interval or region in
the Poisson distribution, then the mean and the variance of the Poisson distribution
are both equal to 𝜇

𝐸 𝑋 = 𝜇 and 𝑉𝑎𝑟 𝑋 = 𝜎 2 = 𝜇
NOTE: In a Poisson distribution, only one parameter, 𝜇 is needed to
determine the probability of an event

A cumulative Poisson probability refers to the probability that the Poisson random
variable is greater than some specified lower limit and less than some specified
upper limit.
Example 1:
2% of the output per month of a mass-produced product have faults. What is the
probability that of a sample of 400 taken that 5 will have faults?
Solution:
𝑒 −𝜇 𝜇 𝑥 𝑙𝑒𝑡 ′ 𝑠 𝑎𝑝𝑝𝑙𝑦; 𝑥 = 5
𝑃 𝑋 =
𝑥! 𝑃 𝑋 =
𝑒 −𝜇 𝜇 𝑥
𝑥!
Assuming the Poisson distribution, we have
𝑒 −8 85
𝑛 = 400 , 𝑝 = 0.02 ; 𝜇 = 𝑛𝑝 , = = 𝟎. 𝟎𝟗𝟑
5!
𝜇 = 𝑛𝑝
= 400 ∗ 0.02
𝜇=8
Example 2:
The output from a CNC machine is inspected by taking samples of 60 items. If the
probability of a defective item is 0.0015, determine the probability of the sample
having
 Two defective items
 More than two defective items

To compute 𝜇
We have 𝑛 = 60 and 𝑝 = 0.0015. Thus, assuming a Poisson distribution, we have

𝜇 = 𝑛𝑝
= 60(0.0015)
𝜇 = 0.09
 Two defective items as  The
: probability of there being more than two
defective items is
𝑥 = 2 ; 𝜇 = 0.09

𝑒 −𝜇 𝜇 𝑥
𝑃 𝑋 =
𝑥! 𝑃 𝑋 > 2 = 𝑃 𝑋 = 3 + 𝑃 𝑋 = 4 + ⋯ + 𝑃(𝑋 = 60)
𝑒 −0.09 0.09 2
𝑃 2 = = 3.7 𝑥 10−3
2! 𝑖𝑛𝑠𝑡𝑒𝑎𝑑 𝑢𝑠𝑒 𝑐𝑜𝑚𝑝𝑙𝑒𝑚𝑒𝑛𝑡 𝑟𝑢𝑙𝑒

= .0037 𝑃 𝑋 > 2 = 1 − 𝑃(𝑋 ≤ 2)

= . 𝟑𝟕% 𝑃 𝑋 > 2 = 1 − 𝑃 0 + 𝑃 1 + 𝑃(2)
= 1 − 0.9139 + 0.0823 +0.0037
𝑒 −0.09 0.090
𝑃 0 = = 0.9139
0!
= 1 − 0.9999
𝑒 −0.09 0.091
𝑃 1 = = 0.0823
1!
𝑒 −0.09 0.092
𝑃 𝑋 > 2 = 0.0001 = 𝟎. 𝟎𝟏%
𝑃 2 = = 0.0037
2!
 X 0 1 2 … 59 60
P(X = x) 0.9139 0.0823 0.0037

𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛 𝑔𝑟𝑎𝑝𝑕

 How to find probabilities for a Poisson distribution when you know only its mean.
Example 3:
The average number of homes sold by the We plug these values into the Poisson
Acme Realty company is 2 homes per day. formula as follows:
What is the probability that exactly 3 homes 𝑒 −𝜇 𝜇 𝑥
will be sold tomorrow? 𝑃 𝑋=𝑥 =
𝑥!
Solution: 𝑒 −2 23
This is a Poisson experiment in which 𝑃 𝑋=3 = = 𝟎. 𝟏𝟖𝟎
3!
we know the following:
μ = 2; since 2 homes are sold per day, Thus, the probability of selling 3 homes
on average. tomorrow is 0.180 or 18%
x = 3; since we want to find the likelihood
that 3 homes will be sold tomorrow.
e ≈ 2.71828;
(𝑏𝑢𝑡 𝑤𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑙𝑦 𝑢𝑠𝑒 𝑜𝑢𝑟 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑜𝑟 ′ 𝑠
𝑒 𝑏𝑢𝑡𝑡𝑜𝑛)
Cumulative Poisson Example 4:
There is a 1.5% probability that a machine
will produce a faulty component. What is
the probability that there will be at least
𝑃 𝑋 ≥ 2 = 𝑃 𝑋 = 2 + ⋯ + 𝑃(𝑋 = 100)
two faulty items in a batch of 100?
𝑖𝑛𝑠𝑡𝑒𝑎𝑑 𝑢𝑠𝑒 𝑐𝑜𝑚𝑝𝑙𝑒𝑚𝑒𝑛𝑡 𝑟𝑢𝑙𝑒
Solution:
𝑃 𝑋 ≥ 2 = 1 − 𝑃(𝑋 < 2)
Assuming the Poisson distribution can be
used, we have n = 100 and p =0.015, 𝜇 = ? 𝑃 𝑋 ≥ 2 = 1 − 𝑃 0 + 𝑃(1)
𝜇 = 𝑛𝑝 = 1 − 0.2231 +0.3347
= 100(0.015) = 1 − 0.5578
𝑒 −𝜇 𝜇 𝑥 𝑃 𝑋 ≥ 2 = 𝟎. 𝟒𝟒𝟐𝟐
𝜇 = 1.5 𝑃 𝑋 =
𝑥!
𝑒 −1.5 1.50
𝑃 𝑋=0 = = 0.2231
0!
𝑒 −1.5 1.51
𝑃 𝑋=1 = = 0.3347
1!
Example 5: To solve this problem, we need to find the
Suppose the average number of lions seen probability that tourists will see 0, 1, 2, or 3
on a 1-day safari is 5. What is the probability lions. Thus, we need to calculate the sum of
that tourists will see fewer than four lions on four probabilities
the next 1-day safari? P (x < 3) =P(x=0) + P(x=1) + P(x=2) + P(x=3).
Solution: To compute this sum, we use the Poisson
formula:
This is a Poisson experiment in which we 𝑒 −𝜇 𝜇 𝑥
know the following: 𝑃 𝑋=𝑥 =
𝑥!
μ = 5; since 5 lions are seen per safari, on P (x < 3) = P(x=0) + P(x=1) + P(x=2) + P(x=3).
average. 𝑒 −5 50 𝑒 −5 51 𝑒 −5 52 𝑒 −5 53
=[ ]+[ ]+[ ]+[ ]
𝑋 < 4 ; x = 0, 1, 2, or 3; since we want to 0! 1! 2! 3!
= [ 0.0067] + [ 0.03369] + [ 0.084224] + [ 0.140375]
find the likelihood that tourists will see fewer
than 4 lions; that is, we want the probability P (x < 3) = 0.2650
that they will see 0, 1, 2, or 3 lions
Thus, the probability of seeing at no more than
e≈ 2.71828;(𝑏𝑢𝑡 𝑤𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑙𝑦 𝑢𝑠𝑒 𝑜𝑢𝑟 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑜𝑟′𝑠 𝑒 𝑏𝑢𝑡𝑡𝑜𝑛) 3 lions is 0.2650.
Example 6: 𝜇 = 1.2; 𝑃 𝑋 = 𝑥 =
𝑒 −𝜇 𝜇𝑥
𝑥!
The number of misprints on a page of the
Daily Mercury has a Poisson distribution with  On page four is 2
mean 1.2. Find the probability that the 𝑒 −1.2 1.22
number of errors 𝑃 𝑋=2 = = 0.217
2!
 On page four is 2  On page three is less than 3
 On page three is less than 3 𝑃 𝑋 <3 =𝑃 𝑋 =0 +𝑃 𝑋 =1 +𝑃 𝑋 =2
𝑒 −1.2 1.20 𝑒 −1.2 1.21 𝑒 −1.2 1.22
Solution: = + +
0! 1! 2!
Let X be the number of errors on one page,
= 𝑒 −1.2 1 + 1.2 + 0.72
= 2.92𝑒 −1.2
𝑃 𝑋 < 3 = 𝟎. 𝟖𝟕𝟗𝟓
Example 7: then Leah receives
1
𝜇= 6 calls in 15 minutes, on average.
Leah’s answering machine receives about six 8
telephone calls between 8 a.m. and 10 a.m. So, μ = 0.75 for this problem.
What is the probability that Leah receives Find P (x > 1) → more than one call
more than one call in the next 15 minutes?
𝑃 𝑋 >1 = 1− 𝑥 ≤ 1
Solution:
= 1− 𝑃 𝑋 =0 +𝑃 𝑋 =1
Let X = the number of calls Leah receives in
𝑒 −𝜇 𝜇 𝑥
15 minutes. (The interval of interest is 15 𝑃 𝑋=𝑥 =
1 𝑥!
minutes or hour.) 𝑒 −0.75 0.750 𝑒 −0.75 0.751
4 =1− +
x = 0, 1, 2, 3, … 0! 1!

If Leah receives, on the average, six = 1 − 0.4724 + 0.3543

telephone calls in two hours, and there are Probability that Leah receives
eight 15-minute intervals in two hours, = 1 − 0.8267 more than one telephone call
in the next 15 minutes
𝑃 𝑋 > 1 = 𝟎. 𝟏𝟕𝟑𝟑 is about 0.1733.
 Example 8: Solution:
Text message users receive or send an  How many text messages does a text
average of 41.5 text messages per day. message user receive or send per hour?
 How many text messages does a text Let X = the number of texts that a user
message user receive or send per hour? sends or receives in one hour.
 What is the probability that a text message The average number of texts received per
42.5
user receives or sends two messages per hour ≈ 1.7292 𝑝𝑒𝑟 𝑕𝑜𝑢𝑟 ;
24
hour? 𝝁 = 𝟏. 𝟕𝟐𝟗
 What is the probability that a text message  What is the probability that a text message
user receives or sends more than two user receives or sends two messages per
messages per hour? hour? 𝑒 −𝜇 𝜇 𝑥
𝑃 𝑋=𝑥 =
𝑥!

𝑒 −1.729 1.7292
𝑃 𝑋=2 = = 0.265
2!
 What is the probability that a text message
user receives or sends more than two
messages per hour?
𝑃 𝑋 > 2 = 𝑃 𝑋 = 3 + …𝑃 𝑋 = 𝑛 + 1
𝑖𝑛𝑠𝑡𝑒𝑎𝑑 , 𝑢𝑠𝑒 𝑡𝑕𝑒 𝑜𝑡𝑕𝑒𝑟 𝑓𝑜𝑟𝑚𝑢𝑙𝑎 𝑠𝑖𝑛𝑐𝑒 𝑛 𝑖𝑛𝑓𝑖𝑛𝑖𝑡𝑒
𝑋 >2 =1−𝑃 𝑋 ≤2

=1− 𝑃 𝑋 =0 +𝑃 𝑋 =1 +𝑃 𝑋 =2
𝑒 −1.729 1.7290 𝑒 −1.729 1.7291 𝑒 −1.729 1.7292
= 1− + +
0! 1! 2!

= 1 − 0.7495
𝑃 𝑋 > 2 = 𝟎. 𝟐𝟓𝟎
Example 9:
On May 13, 2020, starting at 4:30PM, the probability of low seismic activity for the next 48 hours in
Alaska was reported as about 1.02%. Use this information for the next 200 days to find the
probability that there will be low seismic activity in TEN of the next 200 days. Use both the Binomial
and Poisson Distribution to calculate the probabilities. Are they close?
(Geological Science) a sudden release of energy in the earth's crust or upper mantle,
usually caused by movement along a fault plane or by volcanic activity and
resulting in the generation of seismic waves which can be destructive
Solution:
Let X = the number of days with low seismic activity
Using the Binomial Distribution
𝑛!
𝑃 𝑛 = 200, 𝑥 = 10 = 𝑝 𝑥 1 − 𝑝 𝑛−𝑥
𝑥! 𝑛 − 𝑥 !
200!
= 0.010210 0.9898 190
10! 200−10 !

= 2.245100431𝑥1016 0.010210 0.9898 190

= 2.736764898𝑥10−4 0.9898 190

= 3.901691173𝑥10−5
𝑃 𝑛 = 200, 𝑥 = 10 = 𝟎. 𝟎𝟎𝟎𝟎𝟑𝟗
 Example 9:
Using the Poisson Distribution On May 13, 2020, starting at 4:30PM, the probability of low seismic
activity for the next 48 hours in Alaska was reported as about
𝑒 −𝜇 𝜇𝑥
𝑃 𝑋 = 10 = = ; 1.02%. Use this information for the next 200 days to find the
𝑥! probability that there will be low seismic activity in TEN of the next
𝑛 = 200 𝑎𝑛𝑑 𝑝 = 0.0102 200 days. Use both the Binomial and Poisson Distribution to
calculate the probabilities. Are they close?
𝜇 = 𝑛𝑝 = 200 0.0102 ≈ 2.04
𝑒 −𝜇 𝜇𝑥 𝑒 −2.04 2.0410
𝑃 𝑋 = 10 = = = 𝟎. 𝟎𝟎𝟎𝟎𝟒𝟓
𝑥! 10!

𝑤𝑒 𝑒𝑥𝑝𝑒𝑐𝑡 𝑡𝑕𝑒 𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑖𝑜𝑛 𝑡𝑜 𝑏𝑒 𝑔𝑜𝑜𝑑 𝑏𝑒𝑐𝑎𝑢𝑠𝑒 𝑛 𝑖𝑠 𝑙𝑎𝑟𝑔𝑒 𝑎𝑛𝑑 𝑝 𝑖𝑠 𝑠𝑚𝑎𝑙𝑙.

Using the Binomial Distribution 𝑃 𝑛 = 200, 𝑥 = 10 = 𝟎. 𝟎𝟎𝟎𝟎𝟑𝟗

Using the Poisson Distribution 𝑃 𝑋 = 10 = 𝟎. 𝟎𝟎𝟎𝟎𝟒𝟓

𝑇𝑕𝑒 𝑟𝑒𝑠𝑢𝑙𝑡𝑠 𝑎𝑟𝑒 𝑐𝑙𝑜𝑠𝑒

− 𝑏𝑜𝑡𝑕 𝑝𝑟𝑜𝑏𝑎𝑙𝑖𝑡𝑖𝑒𝑠 𝑟𝑒𝑝𝑜𝑟𝑡𝑒𝑑 𝑎𝑟𝑒 𝑎𝑙𝑚𝑜𝑠𝑡 0 𝑖𝑠 𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑙𝑦 𝑎𝑛 𝑎𝑐𝑐𝑝𝑒𝑡𝑎𝑏𝑙𝑒 𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑖𝑜𝑛.
Example 10:
A survey of 500 seniors in the Price Business School yields the following information.
75% go straight to work after graduation. 15% go on to work on their MBA. 9% stay to
get a minor in another program. 1% go on to get a Master’s in Finance. What is the
probability that more than 2 seniors go to graduate school for their Master’s in
finance? Use both the Binomial and Poisson Distribution to calculate the
probabilities. Are they close?
𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 = 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑀𝑎𝑠𝑡𝑒𝑟𝑠 𝑖𝑛 𝐹𝑖𝑛𝑎𝑛𝑐𝑒
Solution: Using Binomial Distribution 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 = 𝑝 = 0.01 ; 𝑛 = 500 , 𝑥 > 2
𝑛!
𝑃 𝑛 = 500 , 𝑥 > 2 = 𝑝𝑥 1 − 𝑝 𝑛−𝑥
𝑥! 𝑛 − 𝑥 ! 500!
0! 500−0 !
0.010 0.99 500−0 = 0.0068
=1− 𝑃 0 +𝑃 1 +𝑃 2 500!
1! 500−1 !
0.011 0.99 500−1 = 0.0332
= 1 − 0.0068 + 0.0332+0.0836
500!
0.012 0.99 500−2 = 0.0836
= 1 − 0.1236 2! 500−2 !

= 𝟎. 𝟖𝟕𝟔𝟒 = 𝟖𝟕. 𝟔𝟒%

 Example 10:
Using Poisson Distribution A survey of 500 seniors in the Price Business School yields the
following information. 75% go straight to work after graduation. 15%
go on to work on their MBA. 9% stay to get a minor in another
𝝁 = 𝑛𝑝 = 500 (0.01) = 5 program. 1% go on to get a Master’s in Finance. What is the
𝑒 −𝜇 𝜇𝑥 probability that more than 2 seniors go to graduate school for their
𝑃 𝑋 >𝑥 == ; Master’s in finance? Use both the Binomial and Poisson Distribution
𝑥! to calculate the probabilities. Are they close?

𝑒 −𝜇 𝜇 𝑥
𝑃 𝑋 >2 ==
𝑥!
𝑃 𝑋 >2 =1−𝑃 𝑋 ≤2
=1− 𝑃 𝑋 =0 +𝑃 𝑋 = 1 +𝑃 𝑋 =2 Are they close?
𝑒 −5 50 𝑒 −5 51 𝑒 −5 52
= 1− + + Using Poisson Distribution= 𝟎. 𝟖𝟕𝟓𝟑
0! 1! 2!
Using Binomial Distribution = 𝟎. 𝟖𝟕𝟔𝟒
= 1 − 0.1247
𝑇𝑕𝑒 𝑟𝑒𝑠𝑢𝑙𝑡𝑠 𝑎𝑟𝑒 𝑐𝑙𝑜𝑠𝑒
𝑃 𝑋 > 2 = 𝟎. 𝟖𝟕𝟓𝟑 − 𝑏𝑜𝑡𝑕 𝑝𝑟𝑜𝑏𝑎𝑙𝑖𝑡𝑖𝑒𝑠 𝑟𝑒𝑝𝑜𝑟𝑡𝑒𝑑 𝑎𝑟𝑒 𝑎𝑙𝑚𝑜𝑠𝑡 1
𝑖𝑠 𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑙𝑦 𝑎𝑛 𝑎𝑐𝑐𝑝𝑒𝑡𝑎𝑏𝑙𝑒 𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑖𝑜𝑛.