Unit 3-Standard distributions

Binomial Distribution

Consider a random experiment with the following properties:

1. The experiment consists of a fixed number 𝑛 of Bernoulli trial, i.e. trials
with exactly two results: success or failure.
2. The trials are identical and independent and hence, the probability of
success 𝑝 remains the same in each trial.
3. The random variable 𝑋 denotes the number of successes obtained in 𝑛
trials.

A random variable X follows binomial distribution if it assumes only non-

negative values with p.m.f 𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥 , 𝑥 = 0,1,2, … , 𝑛
where 𝑛 is the number of trials, 𝑝 is the probability of success, 𝑞 = 1 – 𝑝.
We write 𝑋~𝐵(𝑛, 𝑝)

1. Mean = 𝑛𝑝
2. Variance = 𝑛𝑝𝑞 , Standard Deviation = √𝑛𝑝𝑞
3. MGF = 𝑀𝑋 (𝑡) = (𝑞 + 𝑝𝑒 𝑡 )𝑛
4. Additive property of binomial distribution: If X ∼ B(n1,p) and Y ∼ B(n2,
p) are independent random variables, then X + Y ∼ B(n1 + n2,p)
Derive the moment generating function of the binomial distribution and hence
find the mean and variance.

Answer:
𝑛

𝑀𝑋 (𝑡) = 𝐸(𝑒 𝑡𝑋 ) = ∑ 𝑒 𝑡𝑥 𝑝(𝑥)

𝑥=0
𝑛 𝑛

= ∑ 𝑒 𝑡𝑥 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥 = ∑ 𝑒 𝑡𝑥 𝑛𝐶𝑥 (𝑝𝑒 𝑡 )𝑥 𝑞𝑛−𝑥

𝑥=0 𝑥=0

= (𝑞 + 𝑝𝑒 𝑡 )𝑛 , by binomial expansion formula

To find the mean of the binomial distribution:

𝑑 𝑑 𝑛 𝑛−1 𝑛−1
𝑀𝑋 (𝑡) = (𝑞 + 𝑝𝑒𝑡 ) = 𝑛(𝑞 + 𝑝𝑒𝑡 ) 𝑝𝑒𝑡 = 𝑛𝑝𝑒𝑡 (𝑞 + 𝑝𝑒𝑡 )
𝑑𝑡 𝑑𝑡
𝑑 𝑛−1
𝜇1′ = [ 𝑀𝑋 (𝑡)] = [𝑛𝑝𝑒𝑡 (𝑞 + 𝑝𝑒𝑡 ) ]
𝑑𝑡 𝑡=0 𝑡=0
 𝑛−1
= 𝑛𝑝𝑒0 (𝑞 + 𝑝𝑒0 ) = 𝑛𝑝(𝑞 + 𝑝)𝑛−1 = 𝑛𝑝, since 𝑞 + 𝑝 =1

Hence mean = 𝑛𝑝.

To find the variance

𝑑2 𝑑2 𝑡 )𝑛
𝑑
2
𝑀𝑋 (𝑡) = (𝑞 + 𝑝𝑒 = 𝑛𝑝𝑒 𝑡 (𝑞 + 𝑝𝑒 𝑡 )𝑛−1
𝑑𝑡 𝑑𝑡2 𝑑𝑡
= 𝑛𝑝[𝑒 𝑡 (𝑛 − 1)(𝑞 + 𝑝𝑒 𝑡 )𝑛−2 𝑝𝑒 𝑡 + (𝑞 + 𝑝𝑒 𝑡 )𝑛−1 𝑒 𝑡 ]
𝑑2
𝐸(𝑋 2 ) = [ 𝑀𝑋 (𝑡)]
𝑑𝑡 2 𝑡=0

= 𝑛𝑝[𝑒 0 (𝑛 − 1)(𝑞 + 𝑝𝑒 0 )𝑛−2 𝑝𝑒 0 + (𝑞 + 𝑝𝑒 0 )𝑛−1 𝑒 0 ]

= 𝑛𝑝[(𝑛 − 1)(𝑞 + 𝑝)𝑛−2 𝑝 + (𝑞 + 𝑝)𝑛−1 ]

= 𝑛𝑝[(𝑛 − 1)𝑝 + 1], since 𝑞 + 𝑝 = 1

= 𝑛𝑝(𝑛𝑝 − 𝑝 + 1)

= 𝑛𝑝(𝑛𝑝 + 𝑞) = 𝑛2 𝑝2 + 𝑛𝑝𝑞
2
Variance = 𝐸(𝑋 2 ) − (𝐸(𝑋)) = 𝑛2 𝑝2 + 𝑛𝑝𝑞 − (𝑛𝑝)2

= 𝑛2 𝑝2 + 𝑛𝑝𝑞 − 𝑛2 𝑝2 = 𝑛𝑝𝑞

Hence, Mean=𝑛𝑝, Variance=𝑛𝑝𝑞

Problems

1. 10 coins are thrown simultaneously. Find the probability of getting at

least 7 heads.

Answer:

Here success is getting a head. Let 𝑋 denote the number of heads obtained
(i.e., the number of successes). Then 𝑋 follows binomial distribution with 𝑛 =
1 1
10, 𝑝 = ,𝑞 = 1 − 𝑝 =
2 2

𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥

1 𝑥 1 10−𝑥 1 10
10𝐶𝑥 ( ) ( ) = 10𝐶𝑥 ( ) , 𝑥 = 0,1,2, … ,10
2 2 2
Probability of getting at least 7 heads = 𝑃(𝑋 ≥ 7)

= 𝑃(𝑋 = 7) + 𝑃(𝑋 = 8) + 𝑃(𝑋 = 9) + 𝑃(𝑋 = 10)

 1 10 1 10 1 10 1 10
= 10𝐶7 ( ) + 10𝐶8 ( ) + 10𝐶9 ( ) + 10𝐶10 ( )
2 2 2 2
1 10
= ( ) [10𝐶7 + 10𝐶8 + 10𝐶9 + 10𝐶10 ]
2
1 176
= [120 + 45 + 10 + 1] = = 0.1718
210 1024

2. For a Binomial distribution of mean 4 and variance 2, find the probability

of getting (i) atleast 2 successes. (ii) atmost 2 successes (iii) Find P(5 ≤
X ≤ 7)

Answer:

The p.m.f. of the binomial distribution is

𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥 , 𝑥 = 0,1,2, … , 𝑛

Given : mean=𝑛𝑝 = 4

variance=𝑛𝑝𝑞 = 2

𝑛𝑝𝑞 2 1
∴ = ⟹𝑞=
𝑛𝑝 4 2
1 1
Hence 𝑝 = 1 − 𝑞 = 1 − =
2 2

1
To find 𝑛: 𝑛𝑝 = 4 ⟹ 𝑛 ( ) = 4 ⟹ 𝑛 = 8
2
1 1
∴ 𝑝 = ,𝑞 = ,𝑛 = 8
2 2
𝑥 𝑛−𝑥
1 𝑥 1 8−𝑥 1 8
𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑞 = 8𝐶𝑥 ( ) ( ) = 8𝐶𝑥 ( ) , 𝑥 = 0,1,2, … ,8
2 2 2

(i) P(at least 2 successes)=𝑃(𝑋 ≥ 2) = 1 − 𝑃(𝑋 < 2)

= 1 − [𝑃(𝑋 = 0) + 𝑃(𝑋 = 1)]

1 8 1 8 1 8 9 247
= 1 − [8𝐶0 (2) + 8𝐶1 (2) ] = 1 − (2) [1 + 8] = 1 − 256 = 256

(ii) P(atmost 2 successes) =𝑃 ≤ 2) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2)

1 8 1 8 1 8 1 8 37
= 8𝐶0 (2) + 8𝐶1 (2) + 8𝐶2 (2) = (2) [1 + 8 + 28] = 256

(iii) 𝑃(5 ≤ 𝑋 ≤ 7) = 𝑃(𝑋 = 5) + 𝑃(𝑋 = 6) + 𝑃(𝑋 = 7)

 1 8 1 8 1 8
= 8𝐶5 ( ) + 8𝐶6 ( ) + 8𝐶7 ( )
2 2 2

1 8 92
= ( ) [56 + 28 + 8] =
2 256

3. For usual notation with binomial variate X, find p when n = 6 and

9𝑃(𝑥 = 4) = 𝑃 (𝑥 = 2).

Answer:

The p.m.f. of a binomial distribution is

𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥

𝑃(𝑋 = 4) = 6𝐶4 𝑝4 𝑞 2 , since 𝑛 = 6

𝑃(𝑋 = 2) = 6𝐶2 𝑝2 𝑞 4
Given, 9𝑃(𝑥 = 4) = 𝑃 (𝑥 = 2).

9 × 6𝐶4 𝑝4 𝑞2 = 6𝐶2 𝑝2 𝑞4
9𝑝2 = 𝑞 2 , since 6𝐶4 = 6𝐶2
9𝑝2 − 𝑞 2 = 0
9𝑝2 − (1 − 𝑝)2 = 0
9𝑝2 − (1 − 2𝑝 + 𝑝2 ) = 0
8𝑝2 + 2𝑝 − 1 = 0
−2 ± √4 − (4 × 8 × −1) −2 ± √36 −2 ± 6 1 1
∴𝑝= = = =− ,
16 16 16 2 4

1
Now 𝑝 cannot be negative, so 𝑝 =
4
4. 6 dice are thrown 729 times. How many times do you expect atleast 3
dice to show a 5 or a 6.

Answer:
Success is getting a 5 or a 6 with one dice.
p= probability of getting a 5 or a 6 with one dice
2 1
= =
6 3
2
∴𝑞=
3
Let 𝑋 be the random variable denoting the number of successes when 6
dice are thrown once.
 Given 𝑛 = 6

1 𝑥 2 6−𝑥
∴ 𝑃(𝑋 = 𝑥) = 6𝐶𝑥 ( ) ( )
3 3
Probability that 3 dice show a 5 or a 6
= 𝑃 (𝑋 ≥ 3) = 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) + 𝑃(𝑋 = 6)
1 3 2 6−3 1 4 2 6−4 1 5 2 6−5
=6𝐶3 ( ) ( ) + 6𝐶4 ( ) ( ) + 6𝐶5 ( ) ( ) +
3 3 3 3 3 3

1 6 2 6−6
6𝐶6 ( ) ( )
3 3

1 3 2 3 1 4 2 2 1 5 2 1 1 6 2 0
= 6𝐶3 ( ) ( ) + 6𝐶4 ( ) ( ) + 6𝐶5 ( ) ( ) + 6𝐶6 ( ) ( )
3 3 3 3 3 3 3 3
1
= 6 [20(2)3 + 15(2)2 + 6(2) + 1]
3
1
= 6 [160 + 60 + 12 + 1]
3

233 233
= =
36 729

These 6 dice are thrown 729 times. Hence the expected number of times
233
at least three dice show five or six =729 × 729 = 233 times

5. The chance that one of the 10 telephone lines is busy at any instant is
0.2. (i) What is the most probable number of busy lines and what is the
probability of this number ? (ii) What is the probability that all the lines
are busy?

Answer:

Success represents a line being busy. Let 𝑋 be the number of busy lines. Then
𝑋 follows binomial distribution with 𝑛 = 10, 𝑝 = 0.2, 𝑞 = 0.8

Expected number of busy lines=𝑛𝑝 = 10 × 0.2 = 2

Probability that 2 lines are busy = 𝑃(𝑋 = 2) = 10𝐶2 (0.2)2 (0.8)8 = 0.3020

Probability that all are busy = 𝑃(𝑋 = 10) = 10𝐶10 (0.2)10 (0.8)0 = (0.2)10

5
1 3 t
6. A discrete random variable has m.g.f. M X ( t ) =  + e  . Find E(X),
4 4 
Var(X) and P(X = 2).
Answer:

1 3 5
The given m.g.f. is of the form (𝑞 + 𝑝𝑒 𝑡 )𝑛 = ( + 𝑒 𝑡 )
4 4

∴ 𝑋 follows binomial distribution with

3 1
𝑝 = ,𝑞 = ,𝑛 = 5
4 4
3 𝑥 1 5−𝑥
∴ 𝑃(𝑋 = 𝑥) = 5𝐶𝑥 ( ) ( ) , 𝑥 = 0,1,2,3,4,5,6
4 4
3 15
𝐸(𝑋) = 𝑛𝑝 = 5 × =
4 4
3 1 15
Var(𝑋) = 𝑛𝑝𝑞 = 5 × × =
4 4 16

3 2 1 3 90 45
𝑃(𝑋 = 2) = 5𝐶2 ( ) ( ) = = = 0.08789
4 4 1024 512
7. The probability of a man hitting a target is 1/3. How many times must
he fire so that the probability of hitting the target at least once, is more than
90%.

Answer:

Let 𝑋 be the number of times he hits the target.

1 2
𝑝= . ∴𝑞=
3 3

Then 𝑋 follows a binomial distribution with

𝑥 𝑛−𝑥
1 𝑥 2 𝑛−𝑥
𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑞 = 𝑛𝐶𝑥 ( ) ( )
3 3
Given 𝑃(𝑋 ≥ 1) > 0.9

To find 𝑛

1 − 𝑃(𝑋 < 1) > 0.9

−𝑃(𝑋 < 1) > −0.1

∴ 𝑃(𝑋 < 1) < 0.1

∴ 𝑃(𝑋 = 0) < 0.1

1 0 2 𝑛
𝑛𝐶0 ( ) ( ) < 0.1
3 3
 2 𝑛
( ) < 0.1
3
2
𝑛log ( ) < log0.1
3
𝑛 > 5.65, i.e., 𝑛 ≥ 6
∴ He must fire at least 6 times

8. Out of 800 families with 4 children each, how many families would be
expected to have (i)2 boys and 2 girls (ii) at least 1 boy (iii) atmost 2 girls.
Assume equal probabilities for boys and girls.

Answer:

Let us assume that a girl is a success.

Then 𝑋 = number of girls in the family

𝑋 follows binomial distribution with 𝑛 = 4, 𝑝 = 0.5, 𝑞 = 0.5

𝑃(𝑋 = 𝑥) = 𝑛𝐶𝑥 𝑝 𝑥 𝑞 𝑛−𝑥 = 𝑛𝐶𝑥 (0.5)𝑥 (0.5)𝑛−𝑥

(i) Probability of 2 boys and 2 girls = 𝑃(𝑋 = 2) = 4𝐶2 (0.5)2 (0.5)2 = 0.375

Number of families expected to have 2 boys and 2 girls

=800 × 0.375 = 300

(ii) Probability of at least one boy = 𝑃(𝑋 ≥ 1)

= 4𝐶1 (0.5)1 (0.5)3 + 4𝐶2 (0.5)2 (0.5)2 + 4𝐶3 (0.5)3 (0.5)1 + 4𝐶4 (0.5)4 (0.5)0

= 0.9375
Number of families expected to have atleast one boy

=800 × 0.9375 = 750

(iii) Probability of no girl = 𝑃(𝑋 = 0) = 4𝐶0 (0.5)0 (0.5)4 = 0.0625

Number of families expected to have no girl

= 800 × 0.0625 = 50
(iv) Probability of atmost 2 girls =

𝑃(𝑋 ≤ 2) = 4𝐶0 (0.5)0 (0.5)4 + 4𝐶1 (0.5)1 (0.5)3 + 4𝐶2 (0.5)2 (0.5)2 = 0.6875

Number of families expected to have atmost 2 girls=800 × 0.6875 = 550

9. Fit a binomial distribution for the following data
𝑥: 0 1 2 3 4 5 6 Total
𝑓: 5 18 28 12 7 6 4 80

Answer:

∑ 𝑓𝑥 192
Mean =
∑𝑓
= = 2.4
80

For binomial distribution, mean =𝑛𝑝

𝑛𝑝 = 2.4, 𝑛=6
2.4
∴𝑝= = 0.4
6
𝑞 = 1 − 0.4 = 0.6

∴ 𝑃(𝑋 = 𝑥) = 6𝐶𝑥 𝑝𝑥 𝑞𝑛−𝑥 = 6𝐶𝑥 (0.4)𝑥 (0.6)6−𝑥 , 𝑥 = 0,1,2,3,4,5,6

Expected frequency for 𝑋 = 𝑥 is 800 × 6𝐶𝑥 (0.4)𝑥 (0.6)6−𝑥 , 𝑥 = 0,1,2,3,4,5,6

𝑥 Observed 𝑃(𝑋 = 𝑥) Expected Rounded off

frequency frequency frequency

0 5 0.0467 3.7325 4

1 18 0.1866 14.9299 15

2 28 0.3110 24.8832 25

3 12 0.2765 22.1184 22

4 7 0.1382 11.0592 11

5 6 0.0369 2.9491 3

6 4 0.0041 0.3277 0

Total 80 80

Extra problems

1. If X is a random variable following Binomial distribution, with mean 2.4 and

variance 1.44, find P(X ≥ 5) and P(1 < X ≤ 4).
2. The incidence of occupational disease in an industry is such that the workers
have a 20% chance of suffering from it. What is the probability that out of six
workers chosen, at random, four or more will suffer from the disease?

3. If 20% of the bolts produced by a machine are defective, determine the

probability that out of 4 bolts chosen at random, (i) 1, (ii) 0, (iii) atmost 2 bolts
are defective.

4. In a binomial distribution consisting of 5 independent trials, probabilities of

1 and 2 success are 0.4096 and 0.2048 respectively. Find P(X = 3).

5. 10 coins are tossed 1024 times and the following frequencies are observed.
Compare these frequencies with the expected frequencies.

No of heads 0 1 2 3 4 5 6 7 8 9 10

Frequencies 2 10 38 106 188 252 226 128 59 7 3

6. Fit a binomial distribution for the following data

x: 0 1 2 3 4 5 6 7 8 9

f: 3 8 11 15 16 14 12 11 9 1

7. Fit a binomial distribution for the following data

x: 0 1 2 3 4
f: 5 29 36 25 5

Poisson distribution

If X is a discrete random variable that assumes the values 0,1,2,.... such that
𝑒 − 𝜆 𝜆𝑥
the p.m.f is given by 𝑃(𝑋 = 𝑥) = , 𝑥 = 0,1,2,3, … , 𝜆 > 0, then X is said
𝑥!

to follow Poisson distribution with parameter 𝜆.

1. Mean = 𝜆

2. Variance = 𝜆, Standard deviation =√𝜆

3. Poisson distribution is the limiting case of binomial distribution, when

𝑛 → ∞, 𝑝 → 0 but 𝑛𝑝 = 𝜆 is finite.
𝑡
4. MGF = 𝑒 𝜆(𝑒 −1)
5. Rare events usually follow Poisson distribution.

Derive the m.g.f of Poisson distribution and hence find the mean and variance.
∞
𝑡𝑋 ]
𝑀𝑋 (𝑡) = 𝐸[𝑒 = ∑ 𝑒 𝑡𝑥 𝑝(𝑥)
𝑛=0
∞ ∞
𝑒 − 𝜆 𝜆𝑥 (𝜆𝑒 𝑡 )𝑥
= ∑ 𝑒 𝑡𝑥 . = 𝑒 −𝜆 ∑.
𝑥! 𝑥!
𝑛=0 𝑛=0

2 3
−𝜆 𝜆𝑒𝑡 (𝜆𝑒𝑡 ) (𝜆𝑒𝑡 )
= 𝑒 [1 + + + +⋯]
1! 2! 3!

𝑡 𝑡
= 𝑒−𝜆 𝑒𝜆𝑒 = 𝑒𝜆(𝑒 −1)

𝑡
∴ 𝑀𝑋 (𝑡) = 𝑒𝜆(𝑒 −1)

To find mean

𝑑 𝑡
𝑀𝑋 (𝑡) = 𝑒𝜆(𝑒 −1) . 𝜆𝑒𝑡
𝑑𝑡

𝑑 𝑡 0
∴ Mean =𝐸(𝑋) = [𝑑𝑡 𝑀𝑋 (𝑡)] = [𝑒𝜆(𝑒 −1) . 𝜆𝑒𝑡 ]𝑡=0 = 𝑒𝜆(𝑒 −1) . 𝜆𝑒0 = 𝜆
𝑡=0

To find variance

𝑑2 𝑡 𝑡
2
𝑀𝑋 (𝑡) = 𝑒𝜆(𝑒 −1) . 𝜆𝑒𝑡 . 𝜆𝑒𝑡 + 𝑒𝜆(𝑒 −1) . 𝜆𝑒𝑡
𝑑𝑡

𝑑2 𝑡 𝑡
𝐸(𝑋 2 ) = [ 2 𝑀𝑋 (𝑡)] = [𝑒𝜆(𝑒 −1) . 𝜆𝑒𝑡 . 𝜆𝑒𝑡 + 𝑒𝜆(𝑒 −1) . 𝜆𝑒𝑡 ]𝑡=0
𝑑𝑡 𝑡=0

0 −1) 0 −1)
= 𝑒 𝜆(𝑒 . 𝜆𝑒 0 . 𝜆𝑒 0 + 𝑒 𝜆(𝑒 . 𝜆𝑒 0
 = 𝜆2 + 𝜆

2 2 2
∴ Variance = 𝐸(𝑋 2 ) − (𝐸(𝑋)) = 𝜆 + 𝜆 − 𝜆 = 𝜆

1. If X and Y are independent Poisson variates with parameters 𝜆1 𝑎𝑛𝑑 𝜆2

respectively, then X+Y is also a Poisson variate with parameter𝜆1 + 𝜆2 ., ie.

the sum of two Poisson variates is a Poisson variate. ie, Poisson distribution
follows additive property.

Proof :

X is a Poisson variate with parameter 𝜆1

𝑡
∴ 𝑀𝑋 (𝑡) = 𝑒𝜆1(𝑒 −1)

Y is a Poisson variate with parameter 𝜆2

𝑡
∴ 𝑀𝑌 (𝑡) = 𝑒𝜆2(𝑒 −1)

Now 𝑀𝑋+𝑌 (𝑡) = 𝑀𝑋 (𝑡). 𝑀𝑌 (𝑡), since X and Y are independent

𝑡 𝑡
= 𝑒𝜆1(𝑒 −1) . 𝑒𝜆2(𝑒 −1)

𝑡
= 𝑒(𝜆1+𝜆2)(𝑒 −1)

which is the m.g.f. of a Poisson variate with parameter 𝜆1 + 𝜆2 .

∴ X + Y is a Poisson variate with parameter 𝜆1 + 𝜆2 .

2. If X and Y are a Poisson variates with parameters 1 and 2 respectively,

then X-Y is not a Poisson variate.

Proof:

X is a Poisson variate with parameter 𝜆1

 𝑡
∴ 𝑀𝑋 (𝑡) = 𝑒𝜆1(𝑒 −1)

Y is a Poisson variate with parameter 𝜆2

𝑡
∴ 𝑀𝑌 (𝑡) = 𝑒𝜆2(𝑒 −1)

𝑀𝑋−𝑌 (𝑡) = 𝑀𝑋 (𝑡). 𝑀(−𝑌) (𝑡), since X and Y are independent

= 𝑀𝑋 (𝑡). 𝑀𝑌 (−𝑡)

𝑡 −𝑡
= 𝑒𝜆1(𝑒 −1) . 𝑒𝜆2(𝑒 −1)

𝑡
which cannot be expressed in the form 𝑒 𝜆(𝑒 −1) .

∴ X-Y is not a Poisson variate.

Problems:

1. Let X be a random variable following Poisson distribution such that

𝑃(𝑋 = 2) = 9𝑃(𝑋 + 4) + 90𝑃(𝑋 = 6) . Find the mean and standard
deviation of X.
Answer:
X follows Poisson distribution.
𝑥
𝑒− 𝜆 𝜆
∴ 𝑃(𝑋 = 𝑥) =
𝑥!

Given 𝑃(𝑋 = 2) = 9𝑃(𝑋 + 4) + 90𝑃(𝑋 = 6)

𝑒 − 𝜆 𝜆2 𝑒 − 𝜆 𝜆4 𝑒 − 𝜆 𝜆6
= 9. + 90.
2! 4! 6!

𝜆2 𝜆4 𝜆6
= 9. + 90.
2! 4! 6!

1 𝜆2 𝜆4
= 9. + 90.
2 24 720

1 3 𝜆2 𝜆4
= +.
2 8 8
 4 2
∴𝜆 +3𝜆 −4=0
This is a quadratic equation in 𝜆2 .
2 2
(𝜆 + 4) ( 𝜆 − 1) = 0

Solving, we get 𝜆2 = −4, 1

𝜆2 ≠ −4, ∴ 𝜆2 = 1

𝜆 = ±1

𝜆 is non-negative, hence 𝜆 = 1
For a Poisson distribution, mean=variance= 𝜆

Hence, mean = 𝜆 = 1

Variance = 𝜆 = 1

Standard deviation = √𝜆 = 1

2. Find that probability that atmost 5 defective fuses will be found in a box of
200 fuses if experience shows that 2% are defective.

Answer:
Let X denote the number of defective fuses.
2
𝑛 = 200, 𝑝 = = 0.02
100
X follows Poisson distribution with 𝜆 = 𝑛𝑝 = 200 × 0.02 = 4
𝑥
𝑒− 𝜆 𝜆 𝑒− 4 4𝑥
∴ 𝑃(𝑋 = 𝑥) = = , 𝑥 = 0,1,2,3, …
𝑥! 𝑥!

P(Atmost 5 fuses are defective)=𝑃(𝑋 = 0,1,2,3,4,5)

= 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5)

𝑒 − 4 40 𝑒 − 4 41 𝑒 − 4 42 𝑒 − 4 43 𝑒 − 4 44 𝑒 − 4 45
= + + + + +
0! 1! 2! 3! 4! 5!

−4
40 41 42 43 44 45
=𝑒 [ + + + + + ]
0! 1! 2! 3! 4! 5!
643
= 𝑒− 4 × = 0.7851
15
3. A manufacturer of cotter pins knows that 5% of his product is defective. If
he sells pins in boxes of 100 and guarantees that not more than 10 pins will
be defective, what is the approximate probability that a box will fail to meet
the guaranteed quality.
Answer:
Let 𝑋 be the number of defective pins in a box of 100
Given 𝑛 = 100
𝑝=probability of pin being defective = 5%=0.05
Since 𝑛 is large and 𝑝 is small, we can use Poisson distribution
𝜆 = 𝑛𝑝 = 100 × 0.05 = 5
∴ 𝑋 follows Poisson distribution with

𝑒 − 𝜆 𝜆𝑥 𝑒 − 5 5 𝑥
𝑃(𝑋 = 𝑥) = = , 𝑥 = 0,1,2,3, …
𝑥! 𝑥!
Probability that a box will fail to meet the guaranteed quality =𝑃(𝑋 > 10)
𝑒 − 5 5𝑥 5𝑥
=1 − 𝑃(𝑋 ≤ 10) = 1 − ∑10
𝑥=0 = 1 − 𝑒 − 5 ∑10
𝑥=0 = 0.014
𝑥! 𝑥!

4 . In a certain factory manufacturing razor blades, there is small chance of

1/500 for any blade to be defective. The blade are in packets of 10. Use
Poisson distribution to calculate the approximate number of packets containing
(i) no defective (ii) one defective (iii) 2 defective blades respectively in a
consignment of 10,000 packets.
Answer:
1
Given 𝑝 = 500 , 𝑛 = 10

Let 𝑋 be the number of defectives in a packet

1
𝜆 = 𝑛𝑝 = 10 × = 0.02
500
𝑥 𝑥
𝑒− 𝜆 𝜆 𝑒− 0.02 (0.02)
𝑃(𝑋 = 𝑥) = = , 𝑥 = 0,1,2,3, …
𝑥! 𝑥!
(i) Probability of no defective component=𝑃(𝑋 = 0)
0
𝑒− 0.02 (0.02)
= = 0.9802
0!
Number of packets having no defectives = 10000 × 0.9802 = 9802
(ii) Probability of one defective component=𝑃(𝑋 = 1)
 1
𝑒− 0.02 (0.02)
= = 0.0196
1!
Number of packets having one defective = 10000 × 0.0196 = 196
(iii) Probability of two defective components=𝑃(𝑋 = 2)
2
𝑒− 0.02 (0.02)
= = 0.000196
2!
Number of packets having two defectives = 10000 × 0.000196 = 1.96~2

5. 6 coins are tossed 6400 times . Using Poisson distribution, what is the
approximate probability of getting 6 heads 10 times.
Answer:
Let X be the number of times we get 6 heads
1 6
𝑝=probability of getting six heads in one toss of six coins= (2)

1 6
𝜆 = 𝑛𝑝 = 6400 × ( ) = 100
2
X follows Poisson distribution with

𝑒 − 𝜆 𝜆𝑥 𝑒 − 100 100𝑥
𝑃(𝑋 = 𝑥) = = , 𝑥 = 0,1,2,3, …
𝑥! 𝑥!
𝑒− 100 10010
𝑃(𝑋 = 10) = = 1.025 × 10−30
10!

6. In a company, the monthly breakdown of a machine is a random variable

with Poisson distribution, with average 1.8. Find the probability that the
machine will function for a month (i) without breakdown (ii) with exactly one
breakdown (iii) with at least one breakdown.

Answer:
Let X denote the number of breakdowns in a month
X follows Poisson distribution with mean =𝜆 = 1.8
𝑥 𝑥
𝑒− 𝜆 𝜆 𝑒− 1.8 (1.8)
∴ 𝑃(𝑋 = 𝑥) = = , 𝑥 = 0,1,2,3, …
𝑥! 𝑥!
0
𝑒− 1.8 (1.8)
(i) Probability of no breakdown= 𝑃(𝑋 = 0) = = 0.1653
0!
1
𝑒− 1.8 (1.8)
(ii) Probability of exactly one breakdown= 𝑃(𝑋 = 1) = = 0.2975
1!
(iii) Probability of at least one breakdown= 𝑃(𝑋 ≥ 1)
 = 1 − 𝑃(𝑋 = 0) = 1 − 0.1653 = 0.8347

7. The average number of traffic accidents on a certain section of a highway

is two per week. Assume that the number of accidents follows a Poisson
distribution. Find the probability of (i) no accident in a week
(ii) atmost three accidents in a 2 week period.
Solution:
Let X be the number of accidents per week
The average number of accidents per week = 𝜆 = 2
X follows Poisson distribution with
𝑥
𝑒− 𝜆 𝜆 𝑒− 2 2𝑥
𝑃(𝑋 = 𝑥) = = , 𝑥 = 0,1,2,3, …
𝑥! 𝑥!
(i) Probability of no accident in a week = 𝑃(𝑋 = 0)
𝑒 − 2 20
= 0.1353
0!
(ii) Let Y be the number of accidents in a 2-week period.
By additive property of Poisson distribution, the average number of accidents
in a 2-week period = 2 + 2 = 4
𝑒 − 4 4𝑦
Probability of 𝑦 accidents in a 2-week period =
𝑦!

Probability of atmost 3 accidents in a 2-week period =𝑃(𝑌 ≤ 3)

=𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3)

𝑒− 4 40 𝑒− 4 41 𝑒− 4 42 𝑒− 4 43
= + + +
0! 1! 2! 3!
16 64
= 𝑒− 4 [1 + 4 + + ] = 𝑒− 4 × 23.67 = 0.4335
2 6

8. Fit a Poisson distribution to the following data and calculate the expected
frequencies
Deaths 0 1 2 3 4 Total
Frequency 122 60 15 2 1 200

Answer:
∑ 𝑓𝑥 100
Mean =𝑥̅ = ∑𝑓
= 200 = 0.5

∴ 𝜆 = mean = 0.5
 𝑥 𝑥
𝑒− 𝜆 𝜆 𝑒−0.5 (0.5)
𝑃(𝑋 = 𝑥) == = , 𝑥 = 0,1,2,3, 4
𝑥! 𝑥!
𝑥
𝑒−0.5 (0.5)
The expected frequency for 𝑋 = 𝑥 is 200 × , 𝑥 = 0,1,2,3, 4
𝑥!
The expected frequencies are calculated and tabulated below:

𝑥 𝑓 𝑃(𝑋 = 𝑥) Expected frequency Expected frequency

(Rounded off
0 122 0.60653 121.30 121
1 60 0.30327 60.65 61
2 15 0.07581 15.16 15
3 2 0.01263 2.53 3
4 1 0.00157 0.31 0
Total 200 200

Extra questions

1. If X is a Poisson variate with λ = 1.5, find the probability that (i) X = 2 (ii)
X ≤ 3.

2. A manufacturer produces IC chips out of which 1% are defective. Find

the probability that in a box containing 100 chips with no defectives are
found.

3. Fit a Poisson distribution to the following data and calculate the expected
frequencies
x 0 1 2 3 4 Total
f 109 65 22 3 1 200
4. If a Poisson variate X is such that P(X = 1) = 2P(X = 2). Find P(X = 0) and
var(X).

5. A Hospital switch board receives an average of 4 emergency calls in a 10

minute interval. What is the probability that (i) There are at most 2 emergency
calls in 10 minute interval (ii) There are exactly 3 emergency calls in 10 minute
interval.

6. If X is a Poisson random variable such that P(X = 1) = 0.3, P(X = 2) =

0.2, find P(X = 0) and P(X = 3).
7. The number of accidents involving taxi drivers in a city follows Poisson
distribution, with mean 3. Out of 1000 taxi drivers, find approximately the
number of drivers in a certain year with (i) no accidents, (ii) more than 3
accidents.
8. Fit a Poisson distribution to the following data
x 0 1 2 3 4 5 Total
f 142 156 69 27 5 1 400

Normal Distribution (Gaussian distribution)

A normal distribution, is a distribution that occurs naturally in many

situations. For example, the bell curve is seen in tests like the SAT and GRE.
The bulk of students will score the average (C), while smaller numbers of
students will score a B or D. An even smaller percentage of students score an
F or an A. This creates a distribution that resembles a bell . The bell curve is
symmetrical. Half of the data will fall to the left of the mean; half will fall to
the right.

The empirical rule gives us the percentage of data that falls within a certain
number of standard deviations from the mean:
• 68.2% of the data falls within one standard deviation of the mean.
• 95.4% of the data falls within two standard deviations of the mean.
• 99.7% of the data falls within three standard deviations of the mean.

The standard deviation controls the spread of the distribution. A smaller

standard deviation indicates that the data is tightly clustered around
the mean; the normal distribution will be taller. A larger standard deviation
indicates that the data is spread out around the mean; the normal
distribution will be flatter and wider.
Probability density function
A continuous random variable X follows normal distribution (or Gaussian
−
( x −  )2
1
distribution) if its p.d.f is e 2 2 , −   x  , −      ,   0
 2
1. The parameters are  and  .
2.  is the mean and  is the standard deviation of the distribution.
3. The graph of the distribution is bell shaped and is called the normal
probability curve.
4. The curve is symmetrical about the ordinate at x =  .
5. x – axis is an asymptote to the curve.
6. For the normal distribution, mean = median = mode.
7. The normal distribution with mean = 0 and variance = 1, is called the
standard normal distribution and is denoted as N(0,1).
8. The following is the plot of the standard normal probability density
function.

X −
9. If X is a normal variate, then z = is a standard normal variate. The

z2
1 −
p.d.f. of the standard normal variate is  (z ) = e 2
, −   x  ,
2
10. Area under the standard normal curve = 1

11. If a random variable X follows the normal distribution, then we

write:
12. Calculation of various probabilities
Required Shaded region Calculation
probability

P(0  z  1.5) P(0  z  1.5)

 P(z  0.8) P(z  0.8) = 0.5 − P(0  z  0.8)
(Green
area)

P(z  −1.2) P(z  −1.2) = P(z  1.2) = 0.5 − P(0  z  1.2)

P(0.5  z  1.2) P(0.5  z  1.2) = P(0  z  1.2) − P(0  z  0.5)

P(− 1.2  z  0.8) P(− 1.2  z  0.8) = P(0  z  1.2) + P(0  z  0.8)

P(− 1  z  1) P(− 1  z  1) = 2  P(0  z  1)

P(− 1  z  2) P(− 1  z  2) = P(0  z  1) + P(0  z  2)

P(z  0.8) P(z  0.8) = 0.5 + P(0  z  0.8)

P(z  −0.8) P(z  −0.8) = 0.5 + P(0  z  0.8)

Derive the m.g.f of the normal distribution and hence find the mean and
variance.