BINOMIAL DISTRIBUTION

BERNOULLI  An experiment that consists of two complementary

TRIAL outcomes: a “success” or “failure”.
 This experiment is called a Bernoulli Trial.
 P(success) = p
 P(failure) = 1 − p

BINOMIAL  Counts the number of successes when the experiment

RANDOM
is repeated n times.
VARIABLE
 The n experiments are identical, independent Bernoulli
trials.
 A Binomially distributed random variable must satisfy the following properties:

1. The number of trials n must be fixed.

2. The outcomes of each trial can be classified into two complementary
outcomes.
3. Each trial must be identical, i.e. the probability of a success (p) must
remain constant/ the same for all trials.
4. The trials must be independent, i.e. the outcome of one trial must not
affect the outcome of another.
5. The discrete random variable X counts the number of successes out of
the n trials.
If these properties are satisfied, then we can say X follows a binomial distribution:
X~Bin(n, p)
Example

A die is thrown 10 times. Let X be equal to the number of times a ‘6’ appears.
Two complementary outcomes >> a ‘6’ (success) and a ‘1, 2, 3, 4 or 5’ (failure)
n = 10, p = P(success) =
1
6

Therefore, X~Bin(10, )
1
6

Example

Fourteen percent (14%) of flights from a certain airport are delayed. Consider the next
20 independent flights and the number of flights that take off on-time.
Two complementary outcomes >> on-time (success) and delayed (failure)
X = number of on-time flights, n = 20, p = P(success) = 0.86
Therefore, X~Bin(20, 0.86)
 For any value of n, the P.M.F for a binomially distributed random variable is:

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

 Since a binomially distributed random variable is discrete, we can use the usual
formula (from the previous section) to find the mean/expected value and variance
of the random variable. However…
 It has been shown that these formulae reduce to the following for any binomially
distributed random variable:

E(X) = 𝜇 = 𝑛𝑝
(ON FORMULA SHEET)
Var(X) = 𝜎2 = 𝑛𝑝𝑞
 Example

A student answers 10 questions in a multiple-choice test by guessing each answer.

Each question has 5 possible answers, only one of which is correct. What is the
probability that the student chooses:

a) 3 answers correctly?
b) 7 answers correctly?
c) fewer than 3 answers correctly?
d) at least 5 answers correctly?
e) What is the expected number of questions the student answers correctly?
f) What is the variance and standard deviation of the number of questions the
student answers correctly?
 Solution

We can consider a “success” as getting a question right and consider the 10 questions as 10 independent
Bernoulli trials, then X ~ Bin(10 , 0.2) where X is the random variable representing the number of correct

answers, and P(correct answer)= P(success) = = 0.2.

1
5

a) 3 answers correctly?

𝑃 𝑋 = 𝑥 = 𝑛𝐶𝑥 𝑝 𝑥 1 − 𝑝 𝑛−𝑥 How to input into calculator to arrive at above answer:

P(X=3) = 10C3 (0.2)3 (0.8)10-3 = 0.2013 10 (shift) nCr 3 x 0.23 x 0.87 = 0.2013

b) 7 answers correctly?

𝑃 𝑋 = 𝑥 = 𝑛𝐶𝑥 𝑝 𝑥 1 − 𝑝 𝑛−𝑥

P(X=7) = 10C7 (0.2)7 (0.8)10-7 = 0.000786

c) fewer than 3 answers correctly?

𝑃 𝑋 = 𝑥 = 𝑛𝐶𝑥 𝑝 𝑥 1 − 𝑝 𝑛−𝑥

P(X < 3) = P(X ≤ 2)

= P(X = 0) + P(X = 1) + P(X = 2)
= 10C0 (0.2)0 (0.8)10-0 + 10C1 (0.2)1 (0.8)10-1 + 10C2 (0.2)2 (0.8)10-2
= 0.678

d) at least 5 answers correctly?

P(X ≥ 5) = P(X = 5) + P(X = 6) + ... + P(X = 9) + P(X = 10)

= 1 – P(X ≤ 4)
= 1 – [P(X = 0) + P(X = 1) + P(X = 2) + P(X = 3) + P(X = 4)]
= 1 – [10C0 (0.2)0 (0.8)10-0 + 10C1 (0.2)1 (0.8)10-1 + 10C2 (0.2)2 (0.8)10-2 +10C3 (0.2)3 (0.8)10-3

+ 10C4 (0.2)4 (0.8)10-4]

= 0.0328

e) What is the expected number of questions the student answers correctly?

E(X) = 𝑛𝑝 = 10 × 0,2 = 2

f) What is the variance and standard deviation of the number of questions the
student answers correctly?

Var(X) = 𝑛𝑝𝑞 = 10 × 0,2 × 0,8 = 1,6

∴ 𝜎 = 1.6 = 1.26
 CUMULATIVE BINOMIAL DISTRIBUTION TABLES

 Notice that the calculations needed in parts (c) and (d) of the previous example (on
slides 11 and 12) are time consuming. Instead of using the PMF p(x) to solve the
problems, the CDF F(x) can be used.
 Values for the CDF are found in the Cumulative Binomial Distribution tables (Table A).
 There are several tables – one for each different value of n. The first column gives the
value of n while the second column gives the possible values that the random variable X
can take on. The top row gives common values of p.

Note: These tables give cumulative probabilities P(X ≤ x), so situations that involve the
“<”, “>”and “≥” signs must be adjusted so that they are in a form that uses the “≤” sign
i.e. a “less than or equal to” situation.
 Example
n p
1) Suppose X ~ Bin(12 , 0.6). Find the probability that X is less than or equal to 5:
Here we want to find 𝐹 5 = 𝑃(𝑋 ≤ 5).

Step 1: Find the table that has 𝑛 = 12 in the first column.

Step 2: Choose the value 𝑥 = 5 in the second column.

Step 3: Find 𝑝 = 0.6 in the top row.

𝐹(5) = 𝑃(𝑋 ≤ 5) = 0.1582
 n p
2) In the multiple choice test example X ~ Bin(10 , 0.2):

Part (c): What is the probability that the student chooses fewer than 3 answers correctly?

P(X < 3) = P(X ≤ 2) = 0.6778

To find this go to the table with 𝑛 = 10 in the first column. Choose 𝑥 = 2 in the second column
and choose 𝑝 = 0.2 in the top row. Line up the column and row and the value is
𝐹(2) = 𝑃(𝑋 ≤ 2) = 0.67780
Part (d): What is the probability that the student chooses at least 5 answers correctly?

P(X ≥ 5) = 1 – P(X ≤ 4) = 1 – 0.96720 = 0.0328

Example
An engine manufacturing company records for independent components coming out of an
electric repair shop, 16% have an electrode defect, 8% have a circuit defect and the
remaining percentage have no defect.
If 20 components come out of the repair shop, what is the probability that more than 12 have
no defect?
Solution

n = number of components = 20
Let X = number of components that have no defect

p = P(success) = 1 – (0.16 + 0.08) = 0.76

∴ X ~ Bin (20, 0.76)

more than 12 have no defect ∴ P(X > 12) = P(X ≥13)

= 1 – P(X ≤ 12)
Use the Cumulative Binomial Tables to determine P(X ≤ 12), where n = 20; p = 0.76; X = 12

= 1 – P(X ≤ 12)
= 1 – 0.0321
= 0.9679
If 200 components are released from the electric repair shop, what is the expected number
of components that may have a defect?

Expected number, so we are required to calculate the mean – use formula

E(X) = 𝜇 = 𝑛𝑝

n = 200
defective components = 16% and 8%

E(number of defects) = np
= 200 x (0.16 + 0.08) = 48