Prepared by: M.Riaz M.Sc: Statistics I.U.

B (2006-8): Lecturer Statistics (03337368518)- 1

3. Poisson probability distribution

When Probability of success “P” is very small and “n” is very large and then we use
Poisson distribution and its p.d.f given as
e − µ (µ ) x
P( X = x) =
x!
Examples:
i) Number of typing errors per page in the books
ii) Number of persons born blind per year in large city
iii) Telephone calls in a small interval of time
iv) Number of students cheating others in MBBS entry test
Properties of Poisson experiment
i) The result of each trial can be classified as success or failure
ii) Events occur independently in fixed interval of time or space
iii) Probability of occurrence of the event remains equal in the fixed interval of time or
space
iv) Probability that more than one event occurs in small interval of time or space is
negligible
Poisson random variable
The random variable which denotes the number of successes of Poisson experiment is
called Poisson random variable. It is a discrete variable and ranges from zero to infinity
(0 to ∞ )
Properties of Poisson distribution
i) Total probability of Poisson distribution is unity
ii) Mean ,variance and third moment about mean are equal
iii) It is discrete probability distribution
iv) If µ becomes very large then it approaches to normal distribution
v) If “X” and “Y” are independent Poisson random variables with parameters µ and “v”
then “X+Y” is also Poisson distribution with parameter µ + v
Derivation of Poisson distribution
Poisson distribution is a limiting form of binomial distribution when “P” is small and “n”
is very large as n → ∞ and P → 0 .
Proof:
Let “X” be a rasndom variable having a binomial distribution with probability
distribution function.
n
P( X ) = C P x q n − x X:0,1,2,3,…,n
x

n!
P( X ) = P x q n− x
X !(n − X )!
n(n − 1)(n − 2)....(n − x + 1)(n − x)! x n − x
P( X ) = P q
X !(n − X )!
n(n − 1)(n − 2)....(n − x + 1) x n − x
P( X ) = P q (A)
X!
As we know that mean of binomial distribution
µ = np
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 2

µ µ
p= q = 1− p = 1−
n n
Then eq(A) becomes
n(n − 1)(n − 2)....(n − x + 1) µ x µ
P( X ) = ( ) (1 − ) n − x
X! n n
X
µ
P( X ) = [n(n − 1)(n − 2)....(n − x + 1)]( 1X )(1 − µ ) − x (1 − µ ) n
X! n n n
X
µ 1 2 x −1 1 µ µ
P( X ) = n.n(1 − )n(1 − )....n(1 − ) ( X )(1 − ) − x (1 − ) n
X! n n n n n n
X
µ 1 2 x −1 1 µ µ
P( X ) = n.n.n...n ( X − times )(1 − )(1 − )....(1 − ) ( X )(1 − ) − x (1 − ) n
X! n n n n n n
X
µ 1 2 x −1 1 µ µ
P( X ) = n X (1 − )(1 − )....(1 − ) ( X )(1 − ) − x (1 − ) n
X! n n n n n n
X
µ 1 2 x −1 µ µ
P( X ) = (1 − )(1 − )....(1 − ) (1 − ) − x (1 − ) n
X! n n n n n
1 2 x −1 µ
When n → ∞ then (1 − )(1 − )....(1 − )(1 − ) − x → 0
n n n n
X
µ 1 2 x −1 µ µ
lim it n→∞ b( x, n, p) = lim it n→∞ (1 − )(1 − )....(1 − ) (1 − ) − x (1 − ) n
X! n n n n n
µX 1 2 x −1 µ µ
lim it n→∞ b( x, n, p) = lim it n→∞ (1 − )(1 − )....(1 − )(1 − ) − x lim it n→∞ (1 − ) n
X! n n n n n
X
µ 1 2 x −1 µ µ
lim it n→∞ b( x, n, p ) = (1 − )(1 − )....(1 − )(1 − ) − x lim it n→∞ (1 − ) n
X! ∞ ∞ ∞ ∞ n
µ
µX µ n
lim it n→∞ b( x, n, p ) =
X!
[(1 − 0)(1 − 0)....(1 − 0)(1 − 0) ]lim it
−x
n →∞ (1 −
n
) µ

n µ
µX µ
lim it n→∞ b( x, n, p ) =
X!
[(1)(1)....(1)(1) ] lim it
−x
n →∞ (1 −
n
) µ

µX
lim it n→∞ b( x, n, p) =
X!
[e ] −1 µ

µX
lim it n→∞ b( x, n, p ) = e −µ
X!
Required p.d.f of Poisson distribution
Show that total probability of Poisson distribution is one
Proof: Let by definition
∞
Total probabilty = P( X )
X =0
Let “ X → P ( µ ) X : 0,1,2,3,..., ∞
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 3

e − µ (µ ) x
P( X = x) =
x!
e − µ (µ ) x
Total probability =
X =0 x!
(µ ) x
Total probability = e − µ
X =0 x!

(µ)0 (µ)1 (µ) 2

Total probability = e −µ

0!
+
1!
+
2!
+ ... = e−µ eµ = e−µ +µ = e0 = 1 ( )
Hence Proved
Find first four moments about mean of Poisson distribution
Solution: Let by definition
∞
µ1′ = Mean = XP ( X )
X =0
Let “ X → P ( µ ) X : 0,1,2,3,..., ∞
−µ x
e (µ )
P( X = x) =
x!
∞
e −µ (µ ) x
µ1′ = Mean = XP( X ) = X
X =0 X =0 x!
( µ ) x −1 ( µ )
µ1′ = Mean = e − µ X
X =1 X ( X − 1)!
( µ ) x −1
µ1′ = Mean = e − µ µ
X =1 ( X − 1)!

( µ ) 0 ( µ )1 ( µ ) 2
µ1′ = Mean = e − µ µ
0!
+
1!
+
2!
( )
+ ... = µe − µ e µ = µe − µ + µ = µe 0 = µ
∞ ∞
µ 2′ = E ( X 2 ) = E [( X ( X − 1) + X ] = [( X ( X − 1) P( X )] + XP ( X )
X =0 X =0

(µ ) x−2 (µ ) 2 ∞
µ 2′ = E ( X 2 ) = e − µ X ( X − 1) + XP( X )
X =1 X ( X − 1)( X − 2)! X =0
(µ ) x−2
µ 2′ = E ( X 2 ) = e − µ µ 2 +µ
X = 2 ( X − 2)!

( µ ) 0 ( µ )1 ( µ ) 2
µ 2′ = E ( X 2 ) = e − µ µ 2
0!
+
1!
+
2!
( )
+ ... + µ = e − µ µ 2 e µ + µ = µ 2 + µ

µ 3′ = E ( X 3 ) = E [( X ( X − 1)( X − 2) + 3 X ( X − 1) + X ] = E [ X ( X − 1)( X − 2)] + 3E [ X ( X − 1)] + E ( X )

( µ ) x −3
µ 3′ = e − µ µ 3 X ( X − 1)( X − 3) + 3E [ X ( X − 1)] + E ( X )
X =3 X ( X − 1)( X − 2)( X − 3)!
( µ ) x −3
µ 3′ = e − µ µ 3 ) + E [ X ( X − 1)] + E ( X )
X =3 ( X − 3)!
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 4

µ 3′ = µ 3 + 3µ 2 + µ
µ 4′ == E ( X 4 ) = E [( X ( X − 1)( X − 2)( X − 3) + 6 X ( X − 1)( X − 2) + 7 X ( X − 1) + X ]
µ 4′ = E [( X ( X − 1)( X − 2)( X − 3) + 6 X ( X − 1)( X − 2) + 7 X ( X − 1) + X ]
(µ ) x −4
µ 4′ = e − µ µ 4 X ( X − 1)( X − 3) + 6µ 3 + 7 µ 2 + µ
X =4 X ( X − 1)( X − 2)( X − 3)( X − 4)!
x−4
(µ )
µ 4′ = e − µ µ 4 + 6µ 3 + 7 µ 2 + µ
X =4 ( X − 4 )!
µ 4′ = µ 4 + 6 µ 3 + 7 µ 2 + µ
First four moments about mean
µ1 = µ1′ − µ 1′ = µ − µ = 0
µ 2 = µ 2′ − (µ1′ )2 = µ 2 + µ − µ 2 = µ = Var( X )
µ 2 = µ 2′ − (µ1′ )2 = µ 2 + µ − µ 2 = µ = Var( X )
µ 3 = µ 3′ − 3µ1′µ 2′ + 2(µ1′ )3 = µ 3 + 3µ 2 + µ − 3( µ 2 + µ )(µ ) + 2 µ 3 = µ 3 + 3µ 2 + µ − 3µ 3 − 3µ 2 + 2µ 3 = µ
µ 4 = µ 4′ − 4µ 1′µ 3′ + 6( µ1′ ) 2 µ 2′ − 3(µ1′ )4
µ 4 = µ 4 + 6 µ 3 + 7 µ 2 + µ − 4( µ )( µ 3 + 3µ 2 + µ ) + 6( µ ) 2 ( µ 2 + µ ) − 3(µ )4
µ 4 = µ 4 + 6 µ 3 + 7 µ 2 + µ − 4 µ 4 − 12µ 3 − 4 µ 2 + 6µ 4 + 6µ 3 − 3(µ )4 = 3µ 2 + µ
µ 32 µ 2 1
β1 = = =
µ 23 µ 3 µ
µ 3µ 2 + µ 1
β 2 = 42 = 2
= 3+
µ2 µ µ
Derive moment generating function use it to find mean and variance
Proof: Let by definition of m.g.f
∞
M 0 (t ) = E (e tx ) = e tx P ( X )
X =0
Let “ X → P ( µ ) X : 0,1,2,3,..., ∞
−µ x
e (µ )
P ( X = x) =
x!
∞
e − µ ( µ ) x − µ ∞ ( µe t ) x
M 0 (t ) = e tx =e
X =0 x! X =0 x!
( µe t ) x
∞
( µe t ) 0 ( µe t )1 ( µe t ) 2
M 0 (t ) = e − µ
x!
= e −µ
0!
+
1!
+
2!
t t
( ) t
+ ... = e − µ e µe = e − µ + µe = e µ ( e −1)
X =0

Use it to find mean and variance

∂ ∂
µ1′ = Mean = [M 0 (t )]t = 0 = e µ ( e −1) t =0
∂t ∂t
t
[ ]
t ∂
µ1′ = Mean = e µ ( e −1) µ (e t − 1)
∂t t =0

[
µ1′ = Mean = µe µ ( e −1) e t
t
]t =0 = µe µ ( e
0
e = µe µ (1−1) e 0 = µe 0 = µ
−1) 0
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 5

µ1′ = Mean =
∂
∂t
[M 0 (t )]t =0 = ∂ e µ (e −1)
∂t
t
[ ]
t =0

t ∂
µ1′ = Mean = e µ ( e −1) µ (e t − 1)
∂t t =0

µ 2′ = E ( X 2 ) =
∂2
∂t 2
[M 0 (t )]t =0 = ∂ ∂ M 0 (t ) = ∂ µe µ (e −1) e t
∂t ∂t ∂t
t
[ ]
t =0
t =0

∂
∂t
t
[ ]t ∂
∂t
∂
∂t
t
µ 2′ = E ( X 2 ) = µe µ ( e −1) e t t =0 = µe µ ( e −1) e t + µe t e µ ( e −1)
t =0

t t ∂
µ 2′ = E ( X 2 ) = µe µ ( e −1) e t + µe t e µ ( e −1) µ (e t − 1)
∂t t =0

[ t t
µ 2′ = E ( X 2 ) = µe µ ( e −1) e t + µe t e µ ( e −1) µe t ) t =0 ]
) = [µe ]
t t
µ ( e −1) t
µ 2′ = E ( X 2 e + µ 2 e 2t e µ ( e −1) t =0

µ 2′ = E ( X 2 ) = µ + µ 2
Var( X ) = µ 2′ − (µ1′ ) = µ 2 + µ − µ 2 = µ
2

Derive characteristic function of Poisson distribution

Proof: Let by definition of characteristic function
∞
θ (t ) = E (e itx ) = e itx P( X )
X =0
0r
t
θ (t ) = M 0 (it ) As M 0 (t ) = e µ ( e −1)
it
θ x (t ) = e µ ( e −1)

Derive probability generating function of Poisson distribution

Proof: Let by definition of m.g.f
∞
G (t ) = E (θ x ) = θ x P( X )
X =0
Let “ X → P ( µ ) X : 0,1,2,3,..., ∞
e − µ (µ ) x
P ( X = x) =
x!
−µ
∞
e ( µ ) x − µ ∞ ( µθ ) x
G (t ) = θ x =e
X =0 x! X =0 x!
∞
( µθ ) x ( µθ ) 0 ( µθ )1 ( µθ ) 2
θ (t ) = e − µ
x!
= e −µ
0!
+
1!
+
2!
( )
+ ... = e − µ eθµ = e − µ +θµ = e µ (θ −1)
X =0

θ (t ) = e µ (θ −1) Hence the required result

Show that in Poisson distribution or recurrence formula of Poisson distribution
µ
P ( X = x + 1) = P ( x, µ )
x +1
Proof: As we know that
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 6

e − µ (µ ) x
P ( X = x) = (i)
x!
e − µ ( µ ) x +1
P( X = x + 1) = (ii)
( x + 1)!
Dividing eq (ii) by (i)
e − µ ( µ ) x +1 (µ ) x µ ( x)!( µ ) x µ
P ( X = x + 1) ( x + 1)! ( x + 1) x! ( µ ) x ( x + 1) x! ( x)!( µ ) x µ
= −µ = = =
P( X = x) e (µ ) x (µ ) x (µ ) x ( µ ) x ( x + 1) x!
( x)! ( x)! ( x)!
P( X = x + 1) µ
=
P( X = x) ( x + 1)
µ
P( X = x + 1) = P( X = x)
( x + 1)
Hence proved
Show that recurrence relation in Poisson distribution
d
µ r +1 = µ rµ r −1 + µr
dµ
Use it to find first four moments about mean and moments of skewness and kurtosis
Proof: Let we consider
µ r = E ( X − Mean )r
µr = ( X − µ )r P(x)

As X → P ( µ ) then its Probability function given as

e − µ (µ ) x
P( X = x) = And Mean = µ
x!
−µ
r e µX
µr = (X − µ )
X!
Differentiate with respect to P
−µ
d d r e (µ ) x
µr = (X − µ )
dµ dµ x!

e −µ (µ ) x d (µ ) x d −µ
( X − µ )r + (X − µ ) r
e
d x! dµ x! dµ
µr = −µ
dµ (µ ) x d
+ ( X − µ )r e (µ ) x
x! dµ
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 7

e − µ (µ ) x r (µ )
x
r ( X − µ ) (−1) + ( X − µ )
r −1
e − µ (−1)
d x! x!
µr = −µ
dµ r e (µ ) x
+ (X − µ ) X ( µ ) x −1 (1)
x!
−µ −µ
r −1 e (µ ) x r e (µ ) x
− r (X − µ ) − (X − µ )
d x! x!
µr = −µ
dµ r e (µ ) x
+ (X − µ ) X ( µ ) x −1 (1)
x!

−µ −µ
d (µ ) x (µ ) x
µr = − r ( X − µ )r −1 e + ( X − µ )r e (
− 1 + Xµ −1 )
dµ x! x!

−µ −µ
d (µ ) x (µ ) x X
µr = − r ( X − µ )r −1 e + ( X − µ )r e −1
dµ x! x! µ

−µ −µ
d (µ ) x (µ ) x X − µ
µr = − r ( X − µ )r −1 e + ( X − µ )r e
dµ x! x! µ
d e − µ (µ ) x 1 e − µ (µ ) x
µr = − r (X − µ ) r −1
+ (X − µ ) r
(X − µ )
dµ x! µ x!
−µ −µ
d (µ ) x 1 (µ ) x
µr = − r ( X − µ )r −1 e + ( X − µ )r +1 e
dµ x! µ x!
d
µr = − r ( X − µ )r −1 P( X ) + 1 ( X − µ )r +1 P( X )
dµ µ
d 1
µ r = − rµ r −1 + µ r +1
dµ µ
d 1
µ r = −rµ r −1 + µ r +1
dµ µ
d 1
µ r + rµ r −1 = µ r +1
dµ µ
d
µ r +1 = µ rµ r −1 + µr
dµ
Hence proved
Now first four moments about mean
Put r = 0
µ1 = 0 Where µ 0 = 1
Put r = 1
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 8

d
µ1+1 = µ 1µ 0 + µ1 Where µ 0 = 1 and µ1 = 0
dµ
µ 2 = µ [1] = µ
Put r = 2
d
µ 2+1 = µ 2µ1 + µ1 Where µ 2 = µ and µ1 = 0
dµ
d
µ 3 = µ 2(0) + µ = µ (0 + 1) = µ
dµ

Put r = 3
d
µ 3+1 = µ 3µ 2 + µ3 Where µ 2 = µ and µ 3 = µ
dµ
d
µ 4 = µ 3µ + µ = µ (3µ + 1) = 3µ 2 + µ
dµ
µ 32 µ 2 1
β1 = 3 = 3 =
µ2 µ µ
2
µ 3µ + µ 1
β 2 = 42 = 2
= 3+
µ2 µ µ
State and prove reproductive property of Poisson distribution
Solution:
Statement: Let “ X 1 and X 2 ” be the two independent Poisson variate having with
µ1 and µ 2 parameters respectively. Then X 1 + X 2 also follow Poisson distribution with
µ1 + µ 2 parameters.
Proof: As X 1 → P( µ1 ) then its Probability function given as
e − µ1 ( µ1 ) x1
P ( X = x1 ) = And Mean = µ1
x1!
t
And its m.g.f M x1 (t ) = e µ1 ( e −1)

As X 2 → P( µ 2 ) then its Probability function given as

e − µ 2 ( µ 2 ) x2
P( X = x2 ) = And Mean = µ 2
x2 !
t
And its m.g.f M 0 (t ) = e µ 2 ( e −1)
Z = X 1 + X 2 Then m.g.f of “Z”
M Z (t ) = E (e tZ )
M Z (t ) = E (e t ( X 1 + X 2 )
M Z (t ) = E (e tX 1 +tX 2 ) As X 1 and X 2 are independent then
M Z (t ) = E (e tX 1 ) E (e ttX 2 )
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 9

M Z (t ) = M x1 (t ) M x2 (t )
t t t
M Z (t ) = e µ1 ( e −1) e µ2 ( e −1) = e ( µ1 + µ2 )(e −1)
Hence proved it is also Poisson distribution with parameters µ1 + µ 2
If n → ∞ , then Poisson → normality or state and prove asymptotic property of
Poisson distribution.
Proof:
As we know that E ( X ) = µ and Var ( X ) = µ
Now we consider standard normal variate
X −µ
Z=
σ
X −µ
Z=
µ
Let by definition of m.g.f
X −µ tX −tµ tX − tµ − tµ tX
t
tZ µ µ µ µ µ µ
m Z (t ) = E (e ) = E (e ) = E (e ) = E (e e )=e E (e )
− tµ tX

m Z (t ) = e µ E (e µ ) (A)
As we know that m.g.f of Poisson distribution
t
m x (t ) = e µ ( e −1)
t
Replacing t =
µ
t
µ
m x (t ) = e µ ( e −1)
Then (A) becomes
− tµ t
µ
m Z (t ) = e µ e µ ( e −1)
Taking log on both sides
− tµ t − tµ t t

µ µ (e µ
−1) µ µ (e µ
−1) − tµ µ (e µ
−1)
log mZ (t ) = log e e = log e + log e = log e + log e
µ
t
− tµ µ
log mZ (t ) = + µ (e − 1)
µ
0 1 2 3
t t t t
− tµ µ µ µ µ
log mZ (t ) = + µ( + + + + ... − 1)
µ 0! 1! 2! 3!
2 3
− tµ t t t
log mZ (t ) = + µ (1 + + + + ... − 1)
µ µ 2! µ µ µ 3!
− tµ t t2 t3
log mZ (t ) = + µ( + + + ...)
µ µ 2! µ µ µ 3!
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 10

− tµ µt µt 2 µt 3
log mZ (t ) = + + + + ...
µ µ 2! µ µ µ 3!
t2 t3
log mZ (t ) = + + ...
2 µ 3!
As we know µ = nP Then n → ∞ then np → ∞
Applying limit
t2 t3
log mZ (t ) = lim t n→∞ ( + + ...)
2 µ 3!
t2 t3
log mZ (t ) = + lim t n→∞ + lim t n→∞ ...
2 µ 3!
t2
log mZ (t ) = + 0 + 0 + ...
2
t2
log m Z (t ) =
2
Taking antilog on both sides
t2
m Z (t ) = e 2
It is m.g.f of standardized normal distribution
Hence proved n → ∞ , then Poisson → normality

Q.8.33 (b): Suppose that “X” has a Poisson distribution. If P ( X = 1) = 0.3 and
P ( X = 2) = 0.2 then calculate P ( X = 0) and P ( X = 3)
e−µ µ x
Solution: As we know that P ( X = x) =
x!
e−µ µ 1
P( X = 1) = = e −µ µ
1!
0.3 = e − µ µ (i)
e−µ µ 2 e−µ µ 2
P( X = 2) = =
2! 2
0.2 × 2 = e − µ µ 2
0.4 = e − µ µµ (ii)
−µ
Putting the value of e µ = 0.3 then we get µ
0 .4 = 0 .3 µ
0.4
µ= = 1.33
0 .3
Now we find
e −1.33 (1.33) 0
P( X = 0) = =?
0!
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 11

Q.8.33 ©: A random variable “X” has a Poisson distribution such that

P ( X = 2) = 3P ( X = 4) . Find P ( X = 1)
Solution: P ( X = 2) = 3P ( X = 4)
e−µ µ 2 e−µ µ 4
=3
2! 4!
µ2 µ4
=3
2 24
2
µ µ4
=
2 8
2
µ =4
µ=2
e −2 2 1
P( X = 1) = = 0.271
1!
Q.8.40 (b): In a Poisson distribution the first two frequencies were 250 and 160. Find the
frequencies of the next two values of the variable.
Solution: Given that
e−µ µ 0
NP( X = 0) = N = Ne − µ
0!
Ne − µ = 250
e −µ µ 1
NP( X = 1) = N = Ne − µ µ
1!
Ne − µ µ = 160 Then putting the value of Ne − µ = 250 and we get
250 µ = 160
160 250
µ= = 0.64 N = − 0.64 = 474 Now next two frequencies
250 e
e −0.64 (0.64) 2
NP( X = 2) = 474 = 474(0.1079) = 51
2!
e −0.64 (0.64) 3
NP( X = 3) = 474 = (474)(0.023) = 11
3!

Prepared by: Muhammad.Riaz

M.Sc: Statistics (Silver medalist)
I.U.B (2006-08)
Lecturer: NICAAS College
R.Y.K
03337368518
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 12

Poisson process
The Poisson process is a random physical mechanism in which events occur randomly on
a time or distance scale. The formula of Poisson process is
e − λ t (λ t ) x
P( X = x) = Where X = 0,1,2,..., ∞
x!
t = Number of units of time
X = Number of occurrance in t units of time
λ = Average number of occurrance per unit of time
Properties
i) If " λ " is an average number of occurrance per unit of time or space. Then expected
number of occurrence in fixed interval of time or space “t” is λt
ii) The probability of occurrence two or more events in a small interval or time or space
is negligible
iii) Events occurring in independent intervals of time or space are independent
Example.8.21: Telephone calls are being placed through a certain exchange at random
times on the average of four per minute. Assuming a Poisson process, determine the
probability that in a 15-seconds interval, there are 3 or more calls.
Solution:
λ = 4 min ute or 60 sec ond
15
t= sec onds
60
15 e − λ t (λ t ) x
λt = 4 × = 1 P( X = x) = P ( X ≥ 3) = 1 − P ( X < 3)
60 x!
Example.8.22: Flaws in a certain type of drapery material appear on the average of one
in 150 square feet. If we assume the Poisson distribution, find the probability of at most
one flaw in 225 square feet.
Solution:
λ = one in 150 per square feet
225
t= square feet
150
225 e − λ t (λ t ) x
λt = 1 × = 1 .5 P( X = x) = P ( X ≥ 1) = 1 − P ( X < 1)
150 x!
Example:
a) If there are 3 suicides per 30,000 population in a city. Find the probability that in a
given year there are i) no ii) at least 4 suicides in a city of 120000
Solution:
120000
λ = 3 suicide per 30000 t= =4 λt = 3 × 4 = 12
30000
b) Suppose there is an average of 2 suicides per year per 50000 populations. In a city of
100000, find the probability that in a given year there are i) 0 ii) 1 suicide
Solution:
λ = 2 suicide per 50000
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 13

1000000
t= =2 λt = 2 × 2 = 4
50000
c) The reception office at a building receives an average of 4.5 phone calls per half hour.
Find the probability of receiving exactly 6 phone calls at this office during i) half hour ii)
an hour
Solution:
λ = 4.5 per half hour
30
i) t = =1 λt = 4.5 ×1 = 4.5
30
60
ii t = =2 λt = 4.5 × 2 = 9
30
d) A manufacturer of plywood sheets detects 2 flaws per sheet of 200 square feet. Find
the probability of at most one flaw in 300 square feet sheet purchases by a customer.
Solution:
λ = 2 flaw per 200 square feet
300
t= = 1 .5 λt = 2 × 1.5 = 3
200
e) A certain area in a city is, on the average, hot by 6 hurricanes a year. Find the
probability that in the next 6 months that area will be hit by one hurricane.
Solution:
λ = 6 per year or per 12 month
6
t= = 0 .5 λt = 6 × 0.5 = 3.0
12
Q.8.48 (b): Suppose that customers enter a certain shop at the rate of 30 persons an hour.
Using the Poisson distribution, calculate the probability that in a 2-minute interval, no
customer will enter the shop.
Solution:
λ = 30 persons per hour or 60 min ute
2
t= min ute
60
2 e − λ t (λ t ) x e −1 (1) 0
λt = 30 × = 1 P( X = x) = P( X = 0) = = 0.368
60 x! 0!
Q.8.49 (a): Flaws in plywood occur at random with an average of one flaw per 50 square
feet. What is the probability that a 4 feet × 8 feet sheet will have no flaws? At most one
flaw?
b) A doctor receives an average of 3 telephone calls from 9 p.m until 9 a.m. the next
morning. Assuming arrivals of calls are a Poisson process, what is the probability that the
doctor will not be disturbed by a call if she goes to bed at midnight and rises at 6 a.m?
Solution:
a) λ = one flaw per 50 square feet
32
t= per square 50 sequare feet
50
Prepared by: M.Riaz M.Sc: Statistics I.U.B (2006-8): Lecturer Statistics (03337368518)- 14

32
λt = 1 × = 0.64
50
e − λ t (λ t ) x
P( X = x) =
x!
P ( X ≤ 1) = P ( X = 1) + P ( X = 0)
b) λ = 3 calls in 12 hour
6
t=
12
6 e − λ t (λ t ) x
λ t = 3 × = 1 .5 P( X = x) = P ( X = 0) = ?
12 x!
Q.8.50: A computer system in a company has a breakdown once in 25 days, on the
Average. Assuming breakdowns are a Poisson process, what is the probability of?
i) Exactly one breakdown in the next 10 days ii) more than one in next 10 days?
Solution:
i) λ = one in per 25 days
10
t=
25
10 e − λ t (λ t ) x
λ t = 1 × = 0 .4 P( X = x) = P ( X = 1) = ?
25 x!
ii) P( X > 1) = 1 − P ( X ≤ 1) = 1 − [P( X = 1) + P( X = 0)]
Q.8.51: The number of cars passing over a toll bridge during the time interval 10 to 11
a.m. is 300. The cars pass individually and collectively at random. Find the probability
i) Not more than 4 cars will pass during 1-minute interval 10:45 to 10:46
ii) 5 or more cars will pass during the same interval.
Solution:
i) λ = 300 cars in one hour or 60 min ute
1
t=
60
1
λt = 300 × = 5
60
e − λ t (λ t ) x
P( X = x) =
x!
P ( X ≤ 4) = P ( X + 4) + P ( X = 3) + P ( X = 2) + P ( X = 1) + P ( X = 0) = ?
ii)
P( X ≥ 5) = 1 − P( X < 5) = 1 − [P( X + 4) + P( X = 3) + P( X = 2) + P( X = 1) + P( X = 0)] = ?

Prepared by: Muhammad.Riaz

M.Sc: Statistics (Silver medalist)
I.U.B (2006-08)
Lecturer: NICAAS College
R.Y.K
03337368518