Distributions

MO Faheem, Ph.D.
VIT-AP

1 / 39
Binomial Distribution

The Binomial distribution is a discrete distribution. Let A be an event

of the experiment E . If event A happens, then we call it a success,
otherwise, it is a failure. The probability of success is defined as:

P(A) = p

and the probability of failure is:

P(A) = 1 − p = q .

We assume the following conditions:

1 Each trial results in two disjoint outcomes: a success or a failure.
2 The number of trials, denoted as n, is finite.
3 The trials are independent.
4 The probability of success, p, is constant for each trial. 2 / 39
Random variable of Binomial Distribution

Let X be a random variable that represents the number of times an

event occurs out of n trials. The variable X can take values
0, 1, 2, 3, . . . , n. Then, X is called a binomial random variable with
parameters n and p. The binomial distribution is denoted as:

X ∼ B(n, p).

A random variable X follows the binomial distribution if its probability

mass function (PMF) is given by:
à !
n r n−r
P(X = r ) = p q , r = 0, 1, 2, . . . , n
r
where: p is the probability of success, q = 1 − p is the probability of
failure, nr = r !(nn−! r )! is the binomial coefficient.
¡ ¢

3 / 39
Properties of the Binomial Distribution

1 The sum of probabilities in the binomial distribution satisfies:

à !
n n n
p r q n−r = (p + q)n = 1.
X X
P(X = r ) =
r =0 r =0 r
2 Assume that n trials constitute an experiment E . If the experiment
is repeated N times, then the frequency function of the binomial
distribution is given by:
à !
n r n−r
f (r ) = NP(X = r ) = N p q , r = 0, 1, 2, . . . , n.
r
Pn
3 Mean of a binomial distribution: E (X ) = r =0 rp(r ) = Np.
4 Variance of binomial distribution: Var (X ) = Npq , q = 1 − p

4 / 39
Examples of Binomial Distribution
Example 1: Two players A and B play tennis games. Their chances of
winning a game are in the ratio 3:2 respectively. Find A’s chance of
winning at least two games out of four games played.
Solution: Let X be a random variable denoting the number of
successes. The probability that A wins a game is given by:

2 3
p= , q= .
5 5
The probability mass function of X follows a binomial distribution:
à !
4 r 4−r
P(X = r ) = p q , r = 0, 1, 2, 3, 4.
r
Thus, we calculate:

P(X ≥ 2) = 1 − P(X < 2) = 1 − [P(X = 0) + P(X = 1)].

5 / 39
Now, Ã !µ ¶ µ ¶
0 4
4 2 3 81 81
P(X = 0) = = 1×1× = .
0 5 5 625 625
à !µ ¶ µ ¶
1 3
4 2 3 2 27 216
P(X = 1) = = 4× × = .
1 5 5 5 125 625
Thus,
81 216 297 328
µ ¶
P(X ≥ 2) = 1 − + = 1− = .
625 625 625 625

6 / 39
Examples

Example 2: A test is conducted, which consists of 20 MCQs (multiple

choices questions), with every MCQ having its four options, out of
which only one is correct. Determine the probability that a person
undertaking that test has answered exactly five questions wrong.
Solution: Given: n = 20, n − r = 5, r = 20 − 5 = 15.
The probability of success (giving a right answer) is: p = 41 .
Thus, the probability of failure (giving a wrong answer) is: q = 1 − p = 43 .
Using the binomial distribution formula:
à !µ ¶ µ ¶
15 5
20 1 3
P(X = 5) = = 0.0000034.
5 4 4

7 / 39
Examples

Example 3: Six dice are thrown 729 times. How many times do you
expect at least three dice to show a five or six?
Sol. n = 6, N = 729
2 1
P(S) = p = =
6 3
2
P(F ) = q =
3
à ! à !µ ¶ µ ¶
n r n−r 6 1 r 2 6−r
P(X = r ) = p(r ) = p q =
r r 3 3
Probability at least 3 dice show a 5 or 6 is p(3) + p(4) + p(5) + p(6)

N × (p(3) + p(4) + p(5) + p(6)) = 729 × (p(3) + p(4) + p(5) + p(6)) = 233.

8 / 39
Examples

Example 4: A manufacturer of metal pistons finds that, on average,

12% of his pistons are rejected because they are either oversized or
undersized. Determine the probability that a batch of 10 pistons will
contain (a) No more than two rejects and (b) At least two rejects.
Sol (a). Let X be the number of rejected pistons (where success means
rejection). Given: n = 10, p = 0.12, q = 1 − p = 0.88.
No rejection (X = 0):P(X = 0) = 10 0 10
¡ ¢
0 ¡ (0¢.12) (0.88) = 0.2785.
One rejection (X = 1): P(X = 1) = 10 1 9
1 ¢(0.12) (0.88) = 0.37977.
¡10
Two rejections (X = 2):P(X = 2) = 2 (0.12)2 (0.88)8 = 0.23304.
Thus, the probability of getting no more than 2 rejections is:

P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2) = 0.89131.

9 / 39
Examples

Sol (b).
Instead of calculating for X = 2, 3, 4, . . . , 10, we use:

P(X ≥ 2) = 1 − P(X < 2)

= 1 − (P(X = 0) + P(X = 1))
= 1 − (0.2785 + 0.37977) = 0.34173.

10 / 39
Examples

Example 5: A box contain 12 white balls & 5 red balls. If 4 balls are
drawn at random with replacements. Find the expectation of getting
red balls.
Sol.
5
Probability of getting red ball p = 17 , q = 12
17
E (X ) = 4p = 1.1.
Example 6: 8 coins are tossed at a time, 256 times. Find the expected
frequencies of success (getting head). Also find the mean and S.D.

11 / 39
Poisson distribution

Poisson distribution is a discrete probability distribution and is a limiting

case of the binomial distribution. It can be used under the following
conditions:
1 Each trial results in mutually exclusive outcomes, termed as
success and failure.
2 The number of trials n is very large, that is, n → ∞.
3 The probability of success p is very small, that is, p → 0.
4 The product λ = np, where λ is a positive real number, remains
finite.
Therefore, we express:

λ λ
p= , q = 1− .
n n

12 / 39
Random variable of Poisson distribution

Let X be a discrete random variable taking values 0, 1, 2, . . .. If the

probability mass function is given by

λr e −λ
P(X = r ) = , r = 0, 1, 2, . . . , λ > 0,
r!
then X is said to follow the Poisson distribution with parameter λ. We
denote the Poisson distribution by P(λ), and X is called a Poisson
variate.

13 / 39
Properties of Poisson Distribution

1 Assignment of Probability: The total probability sums to 1, as

λr e −λ
given by ∞
r =0 r ! = 1.
P

2 Distribution Function: The cumulative distribution function

(CDF) is given by
F (x) = P(X ≤ x) = xr=0 P(X = r ) = xr=0 λ re! , x = 0, 1, 2, . . .
P P r −λ

3 Unlike the binomial distribution, the Poisson distribution applies

when events occur randomly in a fixed interval of time or space
rather than as outcomes of a definite number of trials.
4 The binomial probabilities can be approximated using the Poisson
distribution when n is large and p is small, where λ = np.
5 Mean of Poisson distribution E (X ) = λ.
6 Variance of Poisson distribution Var (X ) = λ.

14 / 39
Example of Poisson distribution
Example 1: In a book of 600 pages, there are 60 typographical errors.
Assuming the Poisson law for the number of errors per page, find the
probability that a randomly chosen set of 4 pages will contain no errors.
60
Solution: The average number of errors per page is: λ = 600 = 0.1.
The probability of r errors per page follows the Poisson distribution:

λr e −λ (0.1)r e −0.1
P(X = r ) = = .
r! r!

(0.1)0 e −0.1
P(X = 0) = = e −0.1 ≈ 0.9048.
0!
Since the errors on different pages are independent, the probability that
a random sample of 4 pages contains no errors is:

[P(X = 0)]4 = (0.9048)4 ≈ 0.6703.

15 / 39
Example of Poisson distribution

Example 2: A manufacturer of blades knows that 5% his product is

defective. If he sells blades in boxes of 100 and guarantees that no more
than 0 blades will be defective. What is the probability that a box will
fail to meet the guaranteed quality?
Example 3: In a certain factory turning out optical lenses, there is a
small chance 1/500 for any one lens to be defective. The lenses are
supplied in a pack of 10. Use poison distribution to calculate the
approximate no packets containing no defective, one defective, two
defective, three defective lenses resp. in a consignment of 20,000
packets.

16 / 39
Approximation using Poisson distribution

Example 4: There are 135 students inside a room. The probability that
1
one of the students celebrates his/her birthday today equals to .
365
What is the probability that two or more students are celebrating their
birthday today in that room? Answer this using Poisson distribution.

17 / 39
Approximation using Poisson distribution

Example 4: There are 135 students inside a room. The probability that
1
one of the students celebrates his/her birthday today equals to .
365
What is the probability that two or more students are celebrating their
birthday today in that room? Answer this using Poisson distribution.
1
Solution: X ∼ Binomial(135, ).
365
135 27
Ï The mean of X is np = = = 0.369.
365 73
1
Ï X ∼ Binomial(135 , ) ∼ Pois(0.369).
365

17 / 39
Poisson distribution

£ ¤
P(X ≥ 2) =1 − P(0) + P(1)
" #
e −0.36 e −0.36 × 0.36
=1 − +
1 1
=0.554.

18 / 39
Normal Distribution
The probability density function of the normal random variable X is
given by

(x − µ)2
1 −
f (x) = p e 2σ2 where − ∞ < x < ∞.
σ 2π

19 / 39
Normal Distribution: Properties

Ï All normal curves have the bell shape.

Ï The curve is symmetric with respect to a vertical line x = µ.
Ï The area under the curve is always 1.
Ï The curve is completely determined by the mean µ and the
standard deviation σ.
Ï For the same mean, µ, a smaller value of σ gives a taller and
narrower curve, whereas a larger value of σ gives a flatter curve.

20 / 39
Normal Distribution

σ1 < σ2 .

21 / 39
Normal Distribution

µ1 < µ2 and σ1 < σ2 .

22 / 39
Normal Distribution

Suppose heights at VIT are distributed normally with a mean of 68

inches and a standard deviation of 1.8 inches.

23 / 39
Normal Distribution

Ï How many people are less than 70 inches?

¡ ¢
Ï P X < 70 = 0.8667.

24 / 39
Normal Distribution

Ï How many people are between 67 and 70 inches?

¡ ¢
Ï P 67 < X < 70 = 0.5775.

25 / 39
Normal Distribution

Ï P µ − σ < X < µ + σ ≈ 0.6827.

¡ ¢

Ï P µ − 2σ < X < µ + 2σ ≈ 0.9545.

¡ ¢

Ï P µ − 3σ < X < µ + 3σ ≈ 0.9973.

¡ ¢
26 / 39
Standard Normal Distribution

Since there are a potentially infinite number of Normal Distributions,

usually we calculate using a normalized version, the Standard Normal
Distribution(Z) with mean 0 and standard deviation (and variance) 1:

27 / 39
Standard Normal Distribution

Ï P(−1 < X < 1) = P(X < 1) − P(X < −1) = 0.8413 − 0.1587 = 0.6827.

28 / 39
Standard Normal Distribution

X −µ
Z= .
σ

x2
1 −
f (x) = p e 2 where − ∞ < x < ∞.
2π

29 / 39
Example:
Example: Given a standard normal distribution, find the area under the
curve that lies
(a) To the right of z = 1.84.
(b) Between z = −1.97 and z = 0.86.

(a) 1 − Area to left of z(= 1.84) = 1 − 0.9671 = 0.0329.

(b) Area to left of 0.86−Area to left of(−1.97) = 0.8051−0.0244 ≈ 0.78.
30 / 39
Example:
Example: Given a random variable X having a normal distribution with
µ = 50 and σ = 10, find the probability that X assumes a value between
45 and 62.
45−50
Ï The z values corresponding to 45 and 62 are 10 = −0.5 and
62−50
10 = 1.2.

Ï Area to left of 1.2 − Area to left of (−0.5) = 0.8849 − 0.3085 ≈

0.5764.
31 / 39
Example:

32 / 39
Example:

Example: A soft-drink machine is regulated so that it discharges an

average of 200 milliliters per cup. If the amount of drink is normally
distributed with a standard deviation equal to 15 milliliters, calculate
(a) The probability that the cups contains more than 224 milliliters.
(b) The probability that a cup between 191 and 209 milliliters.
(c) How many cups will probably overflow if 230 milliliter cups are
used for the next 1000 drinks?
(d) The amount of drink in the 25th percentile of the cups.

Ï Let X represent the amount of drink distributed per cup.

Ï The mean is µ = 200 milliliters per cup.
Ï The standard deviation is σ = 15 milliliters.

33 / 39
Example:
(a) We want to compute P(X > 224). We need to convert X into a
standard one to compute the probability.
x − µ 224 − 200
z= = = 1. 6
σ 15
Therefore,

P(X > 224) = P(Z > 1.6)

| {z } | {z }
normal with mean 200 and s.d 15 normal with mean 0 and s.d 1
=1 − P(Z ≤ 1.6)
=1 − 0.9452
=0.0548.

(b) The probability between 191 and 209 milliliters, that is,.
P(191 209 ).
|{z} < X < |{z}
x1 x2
34 / 39
Example:

x1 − µ191 − 200
z1 = = = −0.6
σ 15
x2 − µ 209 − 200
z2 = = = 0. 6
σ 15

P(191 209 ) =P(−

|{z} < X < |{z} 0.6} < Z < |{z}
| {z 0. 6 )
x1 x2 z1 z2

=P(Z < 0.6) − P(Z < −0.6)

=0.7257 − 0.2743
=0.4514.

(c) How many cups will probably overflow if 230 milliliter cups are
used for the next 1000 drinks?
35 / 39
Example:
A cup will overflow if the machine discharges more than 230 milliliters.
The probability that the machine discharges

P(X > 230) =P(Z > 2)

=1 − P(Z ≤ 2)
=1 − 0.9772
=0.0228.

Therefore, the number of cups that will overflow in the next 1000 drinks
is

n · p =1000 × 0.0228
=22.8.
≈23.
36 / 39
Example:

(d) We want to find the amount of drink in the 25th percentile of the
cups.

P(X < α) =0.25

α−µ
=⇒ P(Z < ) =0.25
σ
α−µ
=⇒ = − 0.68
σ
=⇒ α =σ × (−0.68) + µ
=15 × (−0.68) + 200
=189.25.

37 / 39
Example:
Example: The average number of acres burned by forest and range
fires in a large New Mexico county is 4, 300 acres per year, with a
standard deviation of 750 acres. The distribution of the number of
acres burned is normal. What is the probability that between 2, 500 and
4, 200 acres will be burned in any given year?

38 / 39
Example:
Example: The average number of acres burned by forest and range
fires in a large New Mexico county is 4, 300 acres per year, with a
standard deviation of 750 acres. The distribution of the number of
acres burned is normal. What is the probability that between 2, 500 and
4, 200 acres will be burned in any given year?
Solution:
Ï Normal distribution with µ = 4300 and σ = 750.
Ï P(2500 < X < 4200).
2500 − 4300 4200 − 4300
Ï Z= = −2.40 and Z = = −0.13333.
750 750

38 / 39
Example:
Example: The average number of acres burned by forest and range
fires in a large New Mexico county is 4, 300 acres per year, with a
standard deviation of 750 acres. The distribution of the number of
acres burned is normal. What is the probability that between 2, 500 and
4, 200 acres will be burned in any given year?
Solution:
Ï Normal distribution with µ = 4300 and σ = 750.
Ï P(2500 < X < 4200).
2500 − 4300 4200 − 4300
Ï Z= = −2.40 and Z = = −0.13333.
750 750

P(−2.40 < Z < −0.13) =P(Z < −0.13) − P(Z < −2.40)
=0.4483 − 0.0082
=0.4401.
38 / 39
Example:

(ii) What number of burnt acres corresponds to the 38th percentile?

39 / 39
Example:

(ii) What number of burnt acres corresponds to the 38th percentile?

Ï P(X <?) = 0.38 =⇒ P(Z <?) = 0.38 =⇒ Z = −0.31.

39 / 39
Example:

(ii) What number of burnt acres corresponds to the 38th percentile?

Ï P(X <?) = 0.38 =⇒ P(Z <?) = 0.38 =⇒ Z = −0.31.

39 / 39
Example:

(ii) What number of burnt acres corresponds to the 38th percentile?

Ï P(X <?) = 0.38 =⇒ P(Z <?) = 0.38 =⇒ Z = −0.31.

Ï X = 4300 + (−0.31)750 = 4300 − 232.5 = 4067.5.

39 / 39