ISE1204: APPLIED STATISTICS

K. Mutangi

May 5, 2021

K. Mutangi ISE1204: APPLIED STATISTICS

Discrete Probability distributions
Discrete Probability Distributions

Denition
For a discrete random variable, its probability distribution is any
table, graph, or formula that gives each possible value and the
probability of that value.

Example
Consider the experiment of tossing a coin twice and noting the
outcome. The probability distribution for the number of heads is as
follows:
Note that S = {HH , HT , TH , TT }.
Heads (X) 0 1 2
Probability 0.25 0.5 0.25

K. Mutangi ISE1204: APPLIED STATISTICS

Discrete probability distributions
Discrete probability distributions

Example
A fair six sided die is tossed. You win $2 if the result is a 1, you
win $1 if the result is a 6 but otherwise you lose $1. Construct the
probability distribution for your gains or loses. In this case we let
the r.v X= the amount won or lost, then the probability distribution
of X is:
X +2 +1 -1
Probability 1/6 1/6 4/6

Denition
Mean of X: For a discrete random variable X, the mean is denoted
n
E (X ) and is given by E (X ) = xi P (xi )
X

i =1
Where P (xi ) is the probability at xi .
K. Mutangi ISE1204: APPLIED STATISTICS
Discrete Probability Distributions
Discrete Probability Distributions

Denition
Variance: The variance for a r.v is given by
Var (X ) =X E (x − µ)2 = E (x 2 ) − µ2 . For a discrete r.v,
E (x 2 ) = x 2 P (x ), hence the variance is
n
Var (x ) = x12 P (xi ) − µ2 .
X

i =1
Note that µ = E (x ).
Example
Calculate the mean and variance for the r.v X in the game
described in the previous example.

Mean = 2( 61 ) + 1( 16 ) − 1( 46 ) = − 16
E (x 2 ) = 4( 16 ) + 16 + 46 = 96 Hence var (x ) = 9
6 − 1
36 = 53
36
K. Mutangi ISE1204: APPLIED STATISTICS
Bernoulli Distribution
Discrete Probability distributions

In this section we discuss some commonly used discrete

probability distributions, namely bernoulli, binomial, geometric
and Poisson distributions.

Denition
Bernoulli distribution: A random variable X which has two possible
outcomes (say TRUE or FALSE, Head or Tail, success or failure,
etc) is called a Bernoulli random variable. The probability
distribution function (pdf) of X is given by:
fX (x , θ) = θx (1 − θ)1−x , x = 0, 1

The mean of x is E (x ) = θ while the variance is var (x ) = θ(1 − θ).

K. Mutangi ISE1204: APPLIED STATISTICS

 Binomial distribution
Discrete Probability Distributions

Binomial Distribution: When we sum a number of Bernoulli

random variables (say n), we get a binomial random variable
with n trials, a probability of success θ and two possible
outcomes in each trial. The probability of success is the same
from trial to trial.
Denition
Let X = X1 + X2 + . . . + Xn where Xi 's are independent and
identically distributed Bernoulli random variables, then X is called a
binomial random  variable and its pdf is given by:
n

fX (x , n, θ) = θx (1 − θ)n−x , x = 0, 1, 2, . . . , n
x

n
 
= x !(nn−! x )!
x
Mean is E (x ) = nθ and variance is var (x ) = nθ(1 − θ).
K. Mutangi ISE1204: APPLIED STATISTICS
 Binomial distribution
Discrete Probability Distributions

Example
A particular mobile phone link is known to transmit 6% of bits of
information in error. Find the probability that two bits out of the
next 10 transmitted are in error.

The number of trials is 10

Probability of success is 0.06 and that of failure is 0.94
Trials are independent
10
 
P (X = 2) = 0.062 (0.94)8
2

K. Mutangi ISE1204: APPLIED STATISTICS

Geometric Distribution
Discrete Probability distributions

Geometric distribution:
It's the probability that the rst occurrence of success requires
k number of independent trials, each with success probability
θ. If the probability of success on each trial is θ, then the
probability that the kth trial (out of k trials) is the rst success
is P (X = k ) = θ(1 − θ)k −1 , k = 1, 2, 3, ...
The above form of geometric distribution is used for modeling
the number of trials up to and including the rst success.
The geometric distribution is an appropriate model if the
following assumptions are true.
The phenomenon being modelled is a sequence of independent
trials.
There are only two possible outcomes for each trial, often
designated success or failure.
The probability of success, θ is the same for every trial.

K. Mutangi ISE1204: APPLIED STATISTICS

Geometric Distribution
Discrete Probability distributions

Example
Products produced by a machine has a 3% defective rate. What is
the probability that the rst defective occurs in the fth item
inspected?
P [X = 5] = θ[1 − θ]k −1 = 0.03[0.97]4

Example
Let X be a geometric random variable with θ = 0 : 25. What is the
probability that the rst success occurs on the 4th trial?

P [X = 4] = 0.25[0.75]3

K. Mutangi ISE1204: APPLIED STATISTICS

Poisson Distribution

Poisson Distribution A Poisson random variable is a discrete

random variable that describes any random phenomenon for
which a count of some sort is of interest. It takes integer
values from 0 to innity.
Examples of Poisson random variables are as follows:
The number of errors per page that a typist makes.
The number of computer malfunctions in a company per given
period.
The number of particles emitted by a radioactive source in a
given time
The Poisson pdf is given by:
−λ x
P [X = x ] = e x !λ , λ > 0, x = 0, 1, 2, 3, . . .

K. Mutangi ISE1204: APPLIED STATISTICS

Poisson Distribution

Example
Computer malfunctions in a big company occur at an average rate
of 1.5 per hour. What is the probability of observing

[a] 2 malfunctions per hour

[b] more than two malfunctions per hour
[c] at most two malfunctions in two hours
Solutions
[a] P [X = 2] = e 2!1.5
− .
1 5 2

[b] P [X > 2] = 1 − P [X ≤ 2] = 1 − [P [X = 0] + P [X =
1] + P [X = 2]]
[c] Here, the mean changes to λ = 1.5 × 2 = 3 since the time
is now 2 hours. At most 2 means up to and including 2
suggesting that we have
e −3 3 0 e −3 3 2
P [X ≤ 2 ] = + e −3 × 3 +
0! 2!
K. Mutangi ISE1204: APPLIED STATISTICS
Continuous Probability distributions

If a random variable is a continuous variable, its probability

distribution is called a continuous probability distribution.
A continuous probability distribution diers from a discrete
probability distribution in several ways.
The probability that a continuous random variable will assume
a particular value is zero.
As a result, a continuous probability distribution cannot be
expressed in tabular form. A formula is used to describe a
continuous probability distribution.
A continuous pdf has the following properties:
the graph of the density function is continuous over the given
range of values.
The area bounded by the curve of the density function and the
x-axis is equal to 1, when computed over the domain of the
variable.
The probability that a random variable assumes a value
between a and b is equal to the area under the density
function bounded by a and b.
K. Mutangi ISE1204: APPLIED STATISTICS
Normal Probability distribution

The Normal distribution

The normal distribution is characterised by two parameters
only, the mean µ and the variance σ 2 .

Denition
Let X1 , X2 , . . . , Xn be a random sample from a population that is
normally distributed with mean µ and variance σ 2 ., the probability
density function of X is given by

1 (x − µ)2
fX (x , µ, σ 2 ) = √ exp{−[ i 2 ]}, −∞ < x < ∞
2π 2σ

K. Mutangi ISE1204: APPLIED STATISTICS

Normal Probability Distribution

Attributes of the normal distribution:

It is symmetric about the mean i.e
P (X > µ) = P (X < µ) = 0.5
The mean =the mode = median. This follows from (1) since
50% of the observations are below the mean and 50% are
above it (denition of median). The height of the curve
indicates the relative frequency of the observations at each
point. The peak indicates the most frequent value which by
denition is the mode.
68% of the observations fall between the mean minus 1
standard deviation and the mean + 1 standard deviation i.e
P (µ − σ < X < µ + σ) = 0.68, we also have
P (µ − 2σ < X < µ + 2σ) = 0.95. Finally
P (µ − 3σ < X < µ + 3σ) = 0.99.
P (a < X < b)= area under the curve between a and b. The
shape of the normal distribution is bell shaped with mean in
the middle.

K. Mutangi ISE1204: APPLIED STATISTICS

Normal Probability Distribution

Z-Values and the standard normal distribution:

If X ∼ N (µ, σ 2 ) random variable, we dene the transformation
Z = X σ−µ which will be normally distributed with zero mean
and variance of 1 as the standard normal, denoted
Z ∼ N (0, 1).

Example
Let X be a normally distributed random variable with µ = 10 and
standard deviation σ = 2. Find the probability that X lies between
11 and 13.6

K. Mutangi ISE1204: APPLIED STATISTICS

Normal Probability distribution

Solution to Example

11 − 10 13.6 − 10
 
P (11 < x < 13.6) = P <z <
2 2
= P (0.5 < z < 1.80)
= P (z < 1.80) − P (z < 0.5)
= 0.9641 − 0.6915 = 0.2726

Example
The grade point averages of a large population of college students
are approximately normally distributed with mean equal to 2.4 and
standard deviation equal to 0.8. What fraction of the students will
possess a grade point average in excess of 3.0?

K. Mutangi ISE1204: APPLIED STATISTICS

Normal Probability Distribution

Example
Solution:

3 − 2.4
 
P (x > 3.0) = P Z >
0.8
= P (z > 0.75) = 1 − P (z ≤ 0.75)
= 1 − 0.7734 = 0.2266

Theorem
Central limit theorem: If random samples of n observations are
drawn from a population with nite mean, µ, and standard
deviation, σ , then, when n is large, the sample mean x̄ will be
approximately normally distributed with mean µ and standard
deviation √σn . The approximation becomes more and more accurate
as n becomes large.
K. Mutangi ISE1204: APPLIED STATISTICS
Normal Probability Distribution

Theorem
Normal Approximation to the binomial distribution.
When n is suciently large, the binomial variable X will be
approximately normally distributed with mean µ = nθ, and variance
σ 2 = nθ(1 − θ). We may then use areas under a tted normal
curve to approximate binomial probabilities:z = √ x −nθ
nθ(1−θ)

Example
A salesman has found that, on average, the probability of a sale on
a single contact is 0.3. If the salesman contacts 50 customers, what
is the probability that at least 10 will buy?

K. Mutangi ISE1204: APPLIED STATISTICS

Normal Probability Distribution

We use the normal approximation to the binomial since n = 50

is large.
n = 50, θ = 0.3, x ∼ Bin(50, 0.3), var (x ) = 50 ∗ 0.3 ∗ 0.7 = 10.5

9.5 − 15
P (x > 9.5) = P (z > √ )
10.5
= P (z > −1.70) = 1 − P (z ≤ −1.70)
= 1 − 0.0446 = 0.9554

K. Mutangi ISE1204: APPLIED STATISTICS

Exponential Probability distribution

Exponential Distribution.
the exponential distribution (negative exponential distribution)
is the probability distribution that describes the time between
events in a Poisson process, i.e. a process in which events
occur continuously and independently at a constant average
rate.

Denition
A random variable X is described as an exponential r.v if its pdf is
given by λe −λx , λ > 0, 0 < x < ∞

E (x ) = λ1 , var (x ) = λ12
The exponential distribution is used to model the time
between the occurrence of events in an interval of time, or the
distance between events in space.

K. Mutangi ISE1204: APPLIED STATISTICS

Exponential Probability Distribution

The exponential distribution may be useful to model events

such as:
The time between successive failures of a machine
The time from diagnosis until death in patients with
metastatic cancer
The distance between successive breaks in a pipeline
The exponential distribution is an appropriate model if the
following conditions are true.
X is the time (or distance) between events, with X > 0.
The occurrence of one event does not aect the probability
that a second event will occur. That is, events occur
independently.
The rate at which events occur is constant.
Two events cannot occur at exactly the same instant.

K. Mutangi ISE1204: APPLIED STATISTICS

Chisquare Probability distribution

Theorem
Let Z1 , Z2 , . . . , Zn be independent standard normal random
n
variables. Then U = Zi2 is chisquared with n degrees of
X

i =1
freedom.

The chisquare distribution will be used to test amoung other

things the following:
Test of independence of two categorical variables
Test of goodness of t of observed data to hypothetical
distributions

K. Mutangi ISE1204: APPLIED STATISTICS

Student's t Probability distribution
Continuous probability distributions

Theorem
Student's-t distribution
Let Z ∼ N (0, 1) and U ∼ χ2 (r ) be independent variables, then
T = qZU follows a t-distribution with r degrees of freedom.
r

K. Mutangi ISE1204: APPLIED STATISTICS