Chapter 4 RANDOM

VARIABLES

Experiments whose outcomes are

numbers

EXAMPLE:
Select items at random from a batch of size
N until the first defective item is found.
Record the number of non-defective items.
Sample Space: S = {0, 1, 2, . . . , N }

The result from the experiment becomes a

variable; that is, a quantity taking different
values on different occasions. Because the
experiment involves selection at random, we
call it a random variable.

Abbreviation: rv

Notation: capital letter, often X.

REMINDER:
The set of possible values that a random
variable (rv) X can take is called the range
of X.

DEFINITION:
A rv X is said to be discrete if its range
consists of a finite or countable number of
values.

Examples: based on tossing a coin repeatedly

No. of H in 1st 5 tosses: {0, 1, 2, . . . ., 5}
No. of T before first H: {0, 1, 2, . . . .}

Note: although the definition of a discrete rv

allows it to take any finite or countable set of
values, the values are in practice almost
always integers.
DEFINITION:

The Probability Function of a discrete

random variable X is the function p(x)
satisfying

p(x) = Pr(X = x)
for all values x in the range of X.

Abbreviation: pf

Notation: p(x) or pX (x). We use the pX (x)

form when we need to make the identity of
the rv clear.

Terminology: The pf is sometimes given the

alternative name of probability mass function
(pmf).
Properties of a pf

If pX (x) is the pf of a rv X, then

• pX (x) ≥ 0, for all x in the range of X.

•
P
pX (x) = 1, where the sum is taken
over the range of X.

Informal ‘definition’ of a distribution:

The pf of a discrete rv describes how the
total probability, 1, is split, or distributed,
between the various possible values of X.

This ‘split’ or pattern is known as the

distribution of the rv.

Note: The pf is not the only way of

describing the distribution of a discrete rv.
Any 1–1 function of the pf will do.
EXAMPLE: Discrete Uniform Distribution

The rv X is equally likely to take each integer

value in the range 1, 2, . . . , n.

Probability function:

 1 , x = 1, 2, . . . , n ,
pX (x) = n
 0 elsewhere.
Cumulative distribution function:




 0, x < 1,


[x]
FX (x) = n , 1 ≤ x ≤ n,




 1, x ≥ n,


where [x] is the integer part of x.

Note: The cdf is defined for all values of x,

not just the ones in the range of X.
1 for all values
For this distribution, the cdf is n
of x in the range 1 ≤ x < 2, then jumps to n 2,

and so on.
 • It satisfies FX (−∞) = 0 ,
• It satisfies FX (∞) = 1 .

DEFINITION: A random variable is said to

be continuous if its cdf is a continuous
function (see later).
This is an important case, which occurs
frequently in practice.
EXAMPLE: The Exponential Distribution
Consider the rv Y with cdf

 0, y < 0,
FY (y) =
 1 − e−y , y ≥ 0 .

This meets all the requirements above, and is

not a step function.
The cdf is a continuous function.
Mean and Variance

The pf gives a complete description of the

behaviour of a (discrete) random variable. In
practice we often want a more concise
description of its behaviour.
DEFINITION: The mean or expectation of
a discrete rv X, E(X), is defined as
X
E(X) = x Pr(X = x).
x
X
Note: Here (and later) the notation means
x
the sum over all values x in the range of X.
The expectation E(X) is a weighted average
of these values. The weights always sum to 1.
Extension: The concept of expectation can
be generalised; we can define the expectation
of any function of a rv. Thus we obtain, for a
function g(·) of a discrete rv X,
X
E{g(X)} = g(x) Pr(X = x) .
x
Measures of dispersion

There are many possible measures. We look

briefly at three plausible ones.

A. ‘Mean difference’: E{X − E(X)}.

Attractive superficially, but no use.

B. Mean absolute difference: E{|X − E(X)|}.

Hard to manipulate mathematically.

C. Variance: E{X − E(X)}2.

The most frequently-used measure.

Notation for variance: V(X) or Var(X).

That is: V(X) = Var(X) = E{X − E(X)}2.

EXAMPLE:
We find the mean and variance for the
random variable X with pf as in the table:
x 1 2 3 4 5
p(x) = Pr(X = x) 0.1 0.1 0.2 0.4 0.2

P
E(X) = x x Pr(X = x), so

E(X) = (1 × 0.1) + (2 × 0.1) + (3 × 0.2)

+(4 × 0.4) + (5 × 0.2) = 3.5.

Var(X) = E(X 2) − {E(X)}2, and

E(X 2) = (12 × 0.1) + (22 × 0.1) + (32 × 0.2)
+(42 × 0.4) + (52 × 0.2) = 13.7 ,
so Var(X) = 13.7 − (3.5)2 = 1.45.

√
Standard deviation of X: 1.45, or 1.20.
 CONTINUOUS RANDOM
VARIABLES

Introduction

Reminder: a rv is said to be continuous if its

cdf is a continuous function.

If the function FX (x) = Pr(X ≤ x) of x is

continuous, what is Pr(X = x)?

Pr(X = x) = Pr(X ≤ x) − Pr(X < x)

= 0, by continuity
A continuous random variable does not
possess a probability function.

Probability cannot be assigned to individual

values of x; instead, probability is assigned to
intervals. [Strictly, half-open intervals]
Example: We gave earlier an example of a
continuous cdf:

 0, y < 0,
FY (y) =
 1 − e−y , y ≥ 0 .

This is the cdf of what is termed the

exponential distribution with mean 1.

For the case of that distribution, we can find

Pr(Y ≤ 1) = FY (1) = 1 − e−1 = 0.6322

Pr(2 ≤ Y ≤ 3) = FY (3) − FY (2)

= (1 − e−3) − (1 − e−2) = 0.0856

Pr(Y ≥ 2.5) = FY (∞) − FY (2.5)

= 1 − (1 − e−2.5) = 0.0821
The probability density function

Alternative names: pdf,

density function,
density.

Notation for pdf: fX (x)

Recall: The cdf of X is denoted by FX (x)
d FX (x)
Relationship: fX (x) =
dx

Care needed: Make sure f and F cannot be

confused!

Interpretation
• When multiplied by a small number h,
the pdf gives, approximately, the probability
that X lies in a small interval, length h, close
to x.
• If, for example, fX (4) = 2 fX (7), then
X occurs near 4 twice as often as near 7.
Mean and Variance

Reminder: for a discrete rv, the formulae for

mean and variance are based on the
probability function Pr(X = x). We need to
adapt these formulae for use with continuous
random variables.

DEFINITION:
For a continuous rv X with pdf fX (x), the
expectation of a function g(x) is defined as
Z ∞
E{g(X)} = g(x) fX (x) dx
−∞

Hence, for the mean:

Z ∞
E(X) = x fX (x) dx
−∞

Compare this with the equivalent definition

for a discrete random variable:
X X
E(X) = x Pr(X = x) , or E(X) = xpX (x) .
x x
The Uniform Distribution
Distribution of a rv which is equally likely to
take any value in its range, say a to b (b > a).

The pdf is constant:

6
fX (x)

1
b−a

a b
Because fX (x) is constant over [a, b] and
Z ∞ Z b
fX (x) dx = fX (x) dx = 1,
−∞ a

 1 , a < x < b,
b−a

fX (x) =
 0

elsewhere.
Uniform Distribution: Mean and Variance
Z b
1
E(X) = µ = x dx
a b−a
= 1
2 (a + b).

Var(X) = σ 2 = E(X 2) − µ2

(a + b)2
Z b
1
= x2 dx −
a b−a 4

1
= (b − a)2.
12

For example, if a random variable is uniformly

distributed on the range (20,140), then
a = 20 and b = 140, so the mean is 80. The
variance is 1200, so the standard deviation is
34.64.
Properties of the exponential distribution

The distribution has pdf


 λe−λx, x ≥ 0 ,
fX (x) =
 0, x < 0.
and its cdf is given by
Z x
FX (x) = λe−λy dy
0
= 1 − e−λx, x > 0.

Mean and Variance

Z ∞
−λx 1
E(X) = x λe dx = .
0 λ
For the variance, we use integration by parts
to obtain
Z ∞
2
E(X 2) = x2 λe−λx dx = 2 .
0 λ

Hence Var(X) = E(X 2) − {E(X)}2

 2
2 1 1
= 2− = 2.
λ λ λ
Cumulative distribution function
If X ∼ N(µ, σ 2), the cdf of X is the integral:
2
− (x−µ)
Z x
1
FX (x) = √ e 2σ 2 dx.
−∞ σ 2π
This cannot be evaluated analytically.
Numerical integration is necessary: extensive
tables are available.
The Standardised Normal Distribution

The Normal distribution with mean 0 and

variance 1 is known as the standardised
Normal distribution (SND). We usually
denote a random variable with this
distribution by Z. Hence

Z ∼ N(0, 1).
Special notation φ(z) is used for the pdf of
N(0, 1). We write
1 − 1 z2
φ(z) = √ e 2 , −∞ < z < ∞.
2π
The cdf of Z is denoted by Φ(z). We write
Z z
Φ(z) = φ(x) dx
−∞
1 − 1 x2
Z z
= √ e 2 dx
−∞ 2π

Tables of Φ(z) are available in statistical

textbooks and computer programs.