Probability Distributions

In statistics probability distribution yields the possible outcomes for any random events
and it is used as models which are simplified versions of some real-life phenomena.
There are two types of random variables and thus there are two types of probability
distributions, namely discrete and continuous probability distribution.
Discrete Probability Distributions
If a discrete variable can take values with associated probabilities, it is called a discrete
random variable. The values and probabilities are said to form a discrete probability distribution.
There are several types of probability distributions, they are quite referred to very often.
Discrete Uniform Distribution
If a random variable can take on 𝑘 different values with equal probability, we can say that
it has a discrete uniform distribution. Another way of saying discrete uniform distribution would
be a known, finite number of outcomes equally likely to happen.
1
𝑓𝑋 (𝑥; 𝑘) = {𝑘 , 𝑓𝑜𝑟 𝑥 = 1,2, … 𝑘
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

k is a parameter, and it is effect to the shape of a distribution.

(𝑘+1) (𝑘 2 −1)
𝐸(𝑋) = 𝑉(𝑋) =
2 12

Bernoulli Distribution
In an experiment has two possible outcomes “success” and “failure” and their probabilities
are 𝑝 and 1 − 𝑝 respectively. Then the getting a success has a Bernoulli distribution.
𝑝 𝑥 (1 − 𝑝)1−𝑥 , 𝑓𝑜𝑟 𝑥 = 0,1
𝑓𝑋 (𝑥; 𝑝) = {
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
𝑝 is a parameter; where 0 ≤ 𝑝 ≤ 1.
𝐸(𝑋) = 𝑝 𝑉(𝑋) = 𝑝(1 − 𝑝)
Binomial Distribution
Binomial distribution is the discrete probability distribution of the number of successes in
a sequence of independent Bernoulli trials. In general, if the random variable 𝑋 follows the
Binomial distribution with parameters 𝑛 and 𝑝, the probability of getting exactly 𝑥 successes in 𝑛
trials is given by,

(𝑛𝑥)𝑝 𝑥 (1 − 𝑝)𝑛−𝑥 , 𝑓𝑜𝑟 𝑥 = 0,1,2, … , 𝑛

𝑓𝑋 (𝑥; 𝑛, 𝑝) = {
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

𝑛 and 𝑝 are parameters, where 0 ≤ 𝑝 ≤ 1 and 𝑛 ranges over the positive integer.
𝐸(𝑋) = 𝑛𝑝 𝑉(𝑋) = 𝑛𝑝(1 − 𝑝)
Binomial distribution satisfies the following properties,
 There are fixed number of trials.
 The trials are independent.
 Each trial has two possible outcomes, success, and failure.
 The probability of success on each trial does not change from trial to trial.
Example:

1) A die has probability 1⁄6 of coming up 6 when rolled. Let 𝑋 = 1 if the die comes up 6 and
𝑋 = 0 otherwise. What is the distribution of 𝑋?
The success probability is 𝑃(𝑋 = 1) = 1⁄6.
Therefore, 𝑋~𝐵𝑒𝑟𝑛𝑜𝑢𝑙𝑙𝑖(1⁄6)
2) A fair coin is tossed 10 times. Let 𝑋 be the number of heads that appear. What is the
distribution of 𝑋?
There are 10 independent Bernoulli trials each with success probability 𝑝 = 0.5. The
random variable 𝑋 is equal to the number of successes in the 10 trails.
Therefore, 𝑋~𝐵𝑖𝑛(10,0.5).
Geometric Distribution
Assume that a sequence of independent Bernoulli trials is conducted, each with the same
success probability 𝑝. Let 𝑋 represent the number of trials up to and including the first success.
Then 𝑋 is a discrete random variable, which is said to have the Geometric distribution with
parameter 𝑝.
𝑝(1 − 𝑝)𝑥 , 𝑓𝑜𝑟 𝑥 = 0,1,2, …
𝑓𝑋 (𝑥; 𝑝) = {
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Where 𝑥 is number of failure until first success.
𝑞 𝑞
𝐸(𝑋) = 𝑝 𝑉(𝑋) = 𝑝2

Negative Binomial Distribution

The Negative Binomial distribution is an extension of the Geometric distribution. Assume
that independent Bernoulli trials, each with success probability 𝑝, are conducted and let 𝑋 denote
the number of trials up to and including the 𝑟 𝑡ℎ success. Then 𝑋 has the Negative Binomial
distribution with parameters 𝑟 and 𝑝.
𝑥−1 𝑟
(1 − 𝑝)𝑥−𝑟 ,
𝑓𝑋 (𝑥; 𝑟, 𝑝) = { 𝑟 − 1 ) 𝑝
( 𝑓𝑜𝑟 𝑥 = 𝑟, 𝑟 + 1, …
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
𝑟𝑞 𝑟𝑞
𝐸(𝑋) = 𝑉(𝑋) = 𝑝2
𝑝

Example:
1) If the probability is 0.75 that an applicant for a driver’s license will pass the road test on
any given try. What is the probability that an applicant will finally pass the test on the
fourth try?
Let 𝑋 – number of attempts to pass the driving test.
Here 𝑥 = 3 (That is number of times he fails)
𝑝 = 0.75
𝑓𝑋 (3,0.75) = 0.75(1 − 0.75)3 =0.0117
2) In a test of weld strength, 80% of tests result in a fracture in the weld. Let 𝑋 denote the
number of tests up to and including the third beam fracture. What is the distribution of 𝑋?
Find 𝑃(𝑋 = 8).
Since 𝑋 represent the number of trials up to and including the third success, and since the
success probability is 𝑝 = 0.2.
𝑋~𝑁𝐵(3,0.2)
The event 𝑋 = 8 means that the third success occurred on the eight trial. Another way to
say this is that there were exactly two successes in the first 7 trials and the eight trial was a
success. Since all the trials are independent it follows that,
𝑃(𝑋 = 8) = 𝑃(𝑒𝑥𝑎𝑐𝑡𝑙𝑦 𝑡𝑤𝑜 𝑠𝑢𝑐𝑐𝑒𝑠𝑠𝑒𝑠 𝑖𝑛 𝑓𝑖𝑟𝑠𝑡 7 𝑡𝑟𝑖𝑎𝑙𝑠) × 𝑃(𝑆𝑢𝑐𝑐𝑒𝑠𝑠 𝑜𝑛 𝑒𝑖𝑔ℎ𝑡 𝑡𝑟𝑖𝑎𝑙)
 So, number of successes in the first 7 trials has 𝐵𝑖𝑛(7,0.2) distribution. Therefore,
7
𝑃(𝑋 = 8) = ( ) 0.22 0.85 × 0.2 = 0.05505
2
When generalize the results

𝑃(𝑋 = 8) = (72)0.23 0.85 = (𝑥−1

𝑟−1
)𝑝𝑟 (1 − 𝑝)𝑥−𝑟

Hypergeometric Distribution
To obtain the Hypergeometric distribution, Binomial distribution that applies to sampling
without replacement, in which case the trials are not independent.
Let us consider a set of 𝑁 elements of which 𝑀 are looked upon as successes and 𝑁 − 𝑀 as failures.
As in connection with the Binomial distribution, we are interested in the probability of getting 𝑥
successes in 𝑛 trials, but now we are choosing without replacement, 𝑛 of the 𝑁 elements contained
in the set.

(𝑀
𝑥
)(𝑁−𝑀
𝑛−𝑥
)
, 𝑓𝑜𝑟 𝑥 = 0,1,2, … , 𝑛
𝑓𝑋 (𝑥; 𝑁, 𝑀, 𝑛) = { (𝑁)
𝑛
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
𝑁, 𝑀 and 𝑛 are parameters where 𝑛 < 20, and 𝑝 is not constant.
𝑀 𝑀 𝑁−𝑀 𝑁−𝑛
𝐸(𝑋) = 𝑛 𝑁 𝑉(𝑋) = 𝑛 𝑁 ( ) ( 𝑁−1)
𝑁

Example:
As part of an air pollution survey, an inspector decides to examine the exhaust of six of a
company’s 24 trucks. If 4 of the company’s truck’s emit excessive number of pollutants, what is
the probability that none of them will be included in the inspector’s sample?
Let 𝑋-Number of trucks included to the survey
Here 𝑁 = 24, 𝑀 = 4, 𝑛 = 6, 𝑥 = 0

(40)(24−4
6−0
)
𝑃(𝑋 = 0) = 𝑓𝑋 (0; 24,4,6) = = 0.2880
(24
6
)

Poisson Distribution
Which is used to model the number of events occurring within a given time interval. As a
example, the random variable X – Number of telephone calls that arrive during 8-9 am on your
mobile phone.

𝑒 −𝜆 𝜆𝑥
𝑓𝑋 (𝑥; 𝜆) = { 𝑥! , 𝑓𝑜𝑟 𝑥 = 0,1,2, …
0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
λ is a parameter and it is an average number of events in the given time interval, and it is known
as “Shape parameter”.
When 𝑛 → ∞, 𝑝 → 0, 𝑛𝑝 = 𝜆 remains constant.
The Poisson distribution will give good approximation when 𝑛 ≥ 20 and 𝑝 ≤ 0.05 to Binomial
distribution.
𝐸(𝑋) = 𝜆 𝑉(𝑋) = 𝜆
Example:
If 2% of the books bound at a certain bindery have defecting bindings, use the Poisson
approximation to the Binomial distribution to determine the probability that 5 of 400 books bound
by this bindery will have defective bindings.
Let X – Number of defective bindings.
𝑛 = 400 , 𝑥 = 5 , 𝑝 = 0.02
That is 𝜆 = 𝑛𝑝 = 400 × 0.02 = 8
𝑒 −8 85
𝑓𝑋 (5,8) = = 0.093
5!