Module I: Probability Distributions
Topic 2: Binomial Distribution
Dr. P. Rajendra, Professor, Department of Mathematics,
CMRIT, Bengaluru.
Introduction to Binomial Distribution
▶ The Binomial Distribution is a discrete probability
distribution that models the number of successes in a fixed
number of independent Bernoulli trials.
▶ Each trial has two possible outcomes: success (with
probability p) or failure (with probability 1 − p).
▶ The Binomial Distribution is characterized by two parameters:
▶ n: The number of trials.
▶ p: The probability of success on each trial.
Example
In a binary classification problem, the Binomial Distribution can
model the number of correct predictions (successes) out of n total
predictions made by a machine learning model.
Binomial Distribution(Continued..)
Let X be a discrete random variable, p be the probability of
success, and q be the probability of failure. The probability mass
function of the binomial distribution can be defined as:
(
n x n−x
x p q , x ≥0
P(X = x) = b(n, p, x) =
0, Otherwise
where n is the number of trials and n p are the parameters.
The binomial distribution follows the following properties:
1. P(X = x) = b(n, p, x) ≥ 0
2. p
P+ q = 1
n n x n−x
3. x=0 x p q =1
The mean, variance, and standard deviation of the binomial
distribution are given by:
▶ Mean: µ = np
▶ Variance: σ 2 = npq
√
▶ Standard Deviation: σ = npq
Applications in AI and Data Science
▶ Model Evaluation: The Binomial Distribution is used to
model the number of successes (correct predictions) in a fixed
number of trials (predictions).
▶ A/B Testing: In A/B testing, it can model the number of
users who perform a specific action (e.g., click a button) out
of a total number of users exposed to the variant.
▶ Natural Language Processing (NLP): Used to model the
occurrence of a particular word in a fixed number of text
documents.
▶ Spam Detection: Can model the number of spam emails
correctly identified out of a set number of emails.
Derivation for Mean, Variance of a Binomial Random
Variable
Let X be a binomial random variable with parameters n (number
of trials) and p (probability of success). The probability mass
function (PMF) of X is given by:
n x n−x
P(X = x) = p q , x = 0, 1, 2, . . . , n
x
where q = 1 − p.
The mean (expected value) µX of X is defined as:
n
X
µX = E (X ) = x · P(X = x)
x=0
Substituting the PMF:
n
X n x n−x
µX = x p q
x
x=0
We can express this as:
n
X n − 1 x−1 (n−1)−(x−1)
µX = np p q
x −1
x=1
Letting y = x − 1, the sum becomes:
n−1
X n − 1 y (n−1)−y
µX = np p q = np(p + q)n−1
y
y =0
Since p + q = 1, we have: µX = np
To find the variance, we first calculate E (X (X − 1)):
n
X
E (X (X − 1)) = x(x − 1) · P(X = x)
x=0
Substituting the PMF:
n
X n x n−x
E (X (X − 1)) = x(x − 1) p q
x
x=0
This can be simplified as:
n
X n − 2 x−2 (n−2)−(x−2)
2
E (X (X − 1)) = n(n − 1)p p q
x −2
x=2
Letting y = x − 2, the sum becomes:
n−2
2
X n − 2 y (n−2)−y
E (X (X −1)) = n(n−1)p p q = n(n−1)p 2 (p+q)n−2
y
y =0
Since p + q = 1, we have:
E (X (X − 1)) = n(n − 1)p 2
The variance σX2 of X is given by:
σX2 = E (X 2 ) − (E (X ))2
Expanding E (X 2 ):
E (X 2 ) = E (X (X − 1)) + E (X )
Substituting the values:
σX2 = n(n − 1)p 2 + np − (np)2
Simplifying:
σX2 = np(1 − p) = npq
σX2 = npq.
Standard Deviation of the Binomial Distribution
The standard deviation σX is the square root of the variance:
√
σX = npq
Thus, the standard deviation of the binomial distribution is
√
σX = npq.
Problem 1
Let X be a binomially distributed random variable based on 6
repetitions of an experiment with probability of success p = 0.3.
Evaluate the following probabilities: (i)P(X ≤ 3), (ii)P(X > 4)
Solution: The probability mass function of a binomial random
variable X with parameters n and p is given by:
n k
P(X = k) = p (1 − p)n−k
k
where n = 6 and p = 0.3.
(i). Calculate P(X ≤ 3):
P(X ≤ 3) = P(X = 0) + P(X = 1) + P(X = 2) + P(X = 3)
6
P(X = 0) = (0.3)0 (0.7)6 = 1 · 1 · (0.7)6 = (0.7)6
0
6
P(X = 1) = (0.3)1 (0.7)5 = 6 · (0.3) · (0.7)5
1
6
P(X = 2) = (0.3)2 (0.7)4 = 15 · (0.3)2 · (0.7)4
2
6
P(X = 3) = (0.3)3 (0.7)3 = 20 · (0.3)3 · (0.7)3
3
∴ P(X ≤ 3) = (0.7)6 +6·0.3·(0.7)5 +15·(0.3)2 ·(0.7)4 +20·(0.3)3 ·(0.7)3
P(X ≤ 3) ≈ 0.1176 + 0.3025 + 0.3241 + 0.1852 = 0.9294
(ii). Calculate P(X > 4):
P(X > 4) = P(X = 5) + P(X = 6)
6
P(X = 5) = (0.3)5 (0.7)1 = 6 · (0.3)5 · (0.7)
5
6
P(X = 6) = (0.3)6 (0.7)0 = 1 · (0.3)6 · 1
6
∴ P(X > 4) = 6 · (0.3)5 · (0.7) + (0.3)6
P(X > 4) ≈ 0.0102 + 0.0007 = 0.0109
Thus,
P(X ≤ 3) ≈ 0.9294, P(X > 4) ≈ 0.0109.
Problem 2
The probability that a pen manufactured by a company will be
defective is 0.1. If 12 such pens are selected at random, find the
probability that: (i) Exactly 2 pens will be defective, (ii) At most 2
pens will be defective, (iii) None will be defective.
Solution: Given that the probability of a pen being defective is
p = 0.1, and the number of pens selected is n = 12, we use the
binomial distribution formula:
n k
P(X = k) = p (1 − p)n−k
k
where X is the random variable representing the number of
defective pens.
(i). Probability that exactly 2 pens will be defective
12
P(X = 2) = (0.1)2 (0.9)10
2
Calculate 12
2 :
12 12! 12 · 11
= = = 66
2 2!(12 − 2)! 2·1
Then:
P(X = 2) = 66 · (0.1)2 · (0.9)10
P(X = 2) = 66 · 0.01 · 0.3487 = 0.2301
(ii). Probability that at most 2 pens will be defective
P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2)
Calculate each term:
12
P(X = 0) = (0.1)0 (0.9)12 = 1 · 1 · (0.9)12 = (0.9)12
0
P(X = 0) = (0.9)12 = 0.2824
12
P(X = 1) = (0.1)1 (0.9)11 = 12 · 0.1 · (0.9)11
1
P(X = 1) = 0.3766
∴, P(X ≤ 2) = 0.2824 + 0.3766 + 0.2301 = 0.8891
(iii). Probability that none will be defective (P(X = 0))
P(X = 0) = (0.9)12 = 0.2824
Problem 3
The number of telephonic lines busy at an instant is a binomial
variant with a probability of 0.1. If 10 lines are chosen at random,
what is the probability that: (i) No line is busy, (ii) All lines are
busy, (iii) At least one line is busy, (iv) At most two lines are busy.
Solution: Given that the probability of a line being busy is
p = 0.1, and the number of lines chosen is n = 10, we use the
binomial distribution formula:
n k
P(X = k) = p (1 − p)n−k
k
where X is the random variable representing the number of busy
lines.
(i). Probability that no line is busy (P(X = 0))
10
P(X = 0) = (0.1)0 (0.9)10
0
Calculate P(X = 0):
P(X = 0) = 1 · 1 · (0.9)10 = (0.9)10
P(X = 0) ≈ 0.3487
(ii). Probability that all lines are busy (P(X = 10))
10
P(X = 10) = (0.1)10 (0.9)0
10
Calculate P(X = 10):
P(X = 10) = 1 · (0.1)10 · 1 = (0.1)10
P(X = 10) = 10−10 ≈ 10−10 ≈ 0.0000000001
(iii). Probability that at least one line is busy (P(X ≥ 1))
P(X ≥ 1) = 1 − P(X = 0)
Using the value of P(X = 0) calculated earlier:
P(X ≥ 1) = 1 − 0.3487 = 0.6513
(iv). Probability that at most two lines are busy (P(X ≤ 2))
P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2)
Calculate P(X = 1) and P(X = 2):
10
P(X = 1) = (0.1)1 (0.9)9 = 10 · 0.1 · (0.9)9
1
P(X = 1) = 10 · 0.1 · 0.3874 = 0.3874
10
P(X = 2) = (0.1)2 (0.9)8 = 45 · 0.01 · (0.9)8
2
P(X = 2) = 45 · 0.01 · 0.4305 = 0.1937
Add these probabilities:
P(X ≤ 2) = 0.3487 + 0.3874 + 0.1937 = 0.9298
Problem 4
When a coin is tossed 4 times, find the probability of getting: (i)
Exactly one head, (ii) At most three heads, (iii) At least two heads.
Solution: Given that a coin is tossed 4 times, we have a binomial
distribution with n = 4 and probability of getting a head p = 0.5.
The binomial probability formula is:
n k
P(X = k) = p (1 − p)n−k
k
where X is the random variable representing the number of heads.
(i). Probability of getting exactly one head (P(X = 1))
4
P(X = 1) = (0.5)1 (0.5)4−1
1
Calculate P(X = 1):
P(X = 1) = 4 · (0.5)1 · (0.5)3 = 4 · 0.5 · 0.125 = 0.25
(ii). Probability of getting at most three heads (P(X ≤ 3))
P(X ≤ 3) = P(X = 0) + P(X = 1) + P(X = 2) + P(X = 3)
Calculate each term:
4
P(X = 0) = (0.5)0 (0.5)4 = 1 · 1 · 0.0625 = 0.0625
0
4
P(X = 2) = (0.5)2 (0.5)2 = 6 · 0.25 · 0.25 = 0.375
2
4
P(X = 3) = (0.5)3 (0.5)1 = 4 · 0.125 · 0.5 = 0.25
3
Add these probabilities:
P(X ≤ 3) = 0.0625 + 0.25 + 0.375 + 0.25 = 0.9375
(iii). Probability of getting at least two heads (P(X ≥ 2))
P(X ≥ 2) = 1 − P(X < 2) = 1 − (P(X = 0) + P(X = 1))
Using the values calculated earlier:
P(X < 2) = P(X = 0) + P(X = 1) = 0.0625 + 0.25 = 0.3125
P(X ≥ 2) = 1 − 0.3125 = 0.6875
Assignment Problems
1. The probability of germination of a seed in a packet of seeds
is found to be 0.7. If 10 seeds are taken for experimenting on
germination in a laboratory, find the probability that: (i) 8
seeds germinate. [Ans: 0.2334], (ii) At least 8 seeds
germinate. [Ans: 0.3826], (iii) At most 8 seeds germinate.
[Ans: 0.8508]
2. A communication channel receives independent pulses at the
rate of 12 pulses per microsecond. The probability of
transmission error is 0.001 for each microsecond. Compute
the probability of: (i) No error during a microsecond. [Ans:
0.9880], (ii) 1 error. [Ans: 0.0118], (iii) At least 1 error. [Ans:
0.0120], (iv) 2 errors. [Ans: 0.000065], (v) At most 2 errors.
[Ans: 0.9999]
3. In 800 families with 5 children each, how many families would
be expected to have: (i) 3 boys. [Ans: 250], (ii) 5 girls. [Ans:
25], (iii) At most 2 girls. [Ans: 400], (v) Either 2 or 3 boys.
[Ans: 500], [Note:- Assuming the probability for boys and girls
to be equal.]
