DISCUSSION NOTES

SPECIAL DISCRETE DISTRIBUTION

Binomal Distribution
Bernoull trial or experiment
An experiment that results in only two possible outcomes. One outcome may be labelled a success and the
other a failure.

Examples:
• In tossing a coin, let X = 0 if a tail occurs and X = 1 if a head occurs.

• In tossing a die, let Y = 0 if an odd number occurs and Y = 1 if an even number occurs.
• For an examination, let X = 0 if the student failed and X = 1 if the student passed.

Binomial experiment
A probability experiment consisting of n independent and identical Bernoulli trials.
Properties:

1. There must be n repeated trials.

2. The trials must be independent of one another.
3. Each trial results in either a success or a failure. This means each of n trials is a Bernoulli experiment.

4. The probability of success must remain the same for each trial. This probability of success for each
trial is denoted by p. Thus, the probability of failure is denoted by 1-p. Other books use q for the
probability of failure.

Main concern of binomial experiments: the probability of X—the number of successes out of the n trials.

Binomial distribution
For a discrete random variable X that follows a binomial distribution, denote by X Bin(n,p), the probability
mass function or the probability distribution of X for a binomial experiment is given by the following formula
(
(n Cx )(px )(q n−x ) , x = 0, 1, 2, ..., n
P (X = x) =
0 , otherwise

where nCx is the number of combinations of n items taken x at a time;

p is the probability of a success in each trial;
q is 1 − p: the probability of failure in each trial;
n is the number of independent trials
x is the number of successes x = 0, 1, 2, ..., n; and n − x is the number of failures

Note that
n!
n Cx =
x!(n − x)!

i
Examples
1. A coin is tossed three times. What is the probability of getting exactly 2 heads?

X ∼ Bin(n = 3, p = 1/2)

P (X = x) (n Cx )(px )(q n−x )

=
 
n!
P (X = x) = (px )(q n−x )
x!(n − x)!
   2   3−2 
3! 1 1
P (X = 2) =
2!(3 − 2)! 2 2
P (X = 2) = 3/8 ∼ 0.375

2. If you are answering a 5-item multiple choice quiz with 4 choices per item and with only one correct
answer, what is the probability that you will get:

(a) exactly 3 correct answers?

X ∼ Bin(n = 5, p = 1/4)

3 5−313
   
5!
P (X = 3) =
3!(5 − 3)! 4 4
   3  2
5! 1 3
P (X = 3) =
3!(5 − 3)! 4 4
P (X = 3) = 0.6328

(b) fewer than 2 correct answers?

X ∼ Bin(n = 5, p = 1/4)

P (X < 2) = P (X = 0) + P (X = 1)
   0   5−0     0   5−1 
5! 1 3 5! 1 3
P (X < 2) = +
0!(5 − 0)! 4 4 1!(5 − 1)! 4 4
   0   5     0   4 
5! 1 3 5! 1 3
P (X < 2) = +
0!(5)! 4 4 1!(4)! 4 4
P (X < 2) = 0.6328 ∼ 63.28%

(c) at least 1 correct answer?

X ∼ Bin(n = 5, p = 1/4)

P (X ≥ 1) 1 − P (X = 0)
=
   0   5−0 
5! 1 3
P (X ≥ 1) = 1 −
0!(5 − 0)! 4 4
   0  5
5! 1 3
P (X ≥ 1) =
0!(5 − 0)! 4 4
P (X ≥ 1) = 0.7627 ∼ 76.27%

ii
(d) at least 2 correct answers?
X ∼ Bin(n = 5, p = 1/4)

P (X ≥ 2) 1 − [P (X = 0) + P (X = 1)]
=
   0   5−0     0   5−1 
5! 1 3 5! 1 3
P (X ≥ 2) = 1 − +
0!(5 − 0)! 4 4 1!(5 − 1)! 4 4
   0   5     0   4 
5! 1 3 5! 1 3
P (X ≥ 2) = 1 − +
0!(5)! 4 4 1!(4)! 4 4
P (X ≥ 2) = 0.3672 ∼ 36.72%

(e) find the expected number of correct answers and the variance

X ∼ Bin(n = 5, p = 1/4)

For a binomial distribution,

E[X] = µx = np; and

V ar[X] = np(1 − p) ∼ npq (since 1-p = q)

Then,
 
1 1
E[X] = µx = 5 = 1 ; and
4 4
  
1 3 15
V ar[X] = 5 = ∼ 0.9375 ∼ 93.75%
4 4 16

iii