Random variables: expected value,
variance and covariance
Definition of some basic terms in
probability
• In statistics, any process of observation or measurement
is called an experiment.
• All statistical experiments have three things in common:
– The experiment has more than one possible
outcome.
– Each possible outcome can be specified in
advance.
– The outcome of the experiment depends on chance.
• A coin toss has all the attributes of a statistical
experiment.
• There is more than one possible outcome.
• We can specify each possible outcome (i.e., heads or
tails) in advance.
• And there is an element of chance (since the outcome is
uncertain).
• The result that we obtain from an experiment is called an
outcome.
• For instance, rolling a die is an experiment.
• It is a well-defined action that leads to a well-defined
result or outcome (Note: In rolling a die the outcomes are
the number of points that face up).
• That well-defined outcome is either of the numbers 1, 2,
3, 4, 5 or 6.
• The set of all possible outcomes for an experiment is
called the sample space (denoted by S).
• In an experiment of rolling a die once, the sample space
is S = {1, 2, 3, 4, 5, 6}.
• The sample space for an experiment of tossing a coin
once is S = {Head, Tail}.
Random variables
• In many experiments, the outcomes may be non-
numerical.
• In assessing the quality of products, we may record the
outcomes as ‘defective’ and ‘non-defective’.
• In classifying medical test results for a certain infectious
disease, we may use the categories ‘positive’ and
‘negative.
• In analyzing the returns of investments, we may record
the outcomes as ‘low’ and ‘high’.
• However, our task gets much easier by quantifying the
outcomes of a random process so that we can perform
any mathematical computation of interest.
• For this purpose, we introduce the concept of a random
variable.
• A random variable is a function or rule that assigns
numerical values to each simple event of a sample
space.
• In other words, it is a function that performs a mapping
of the outcomes of a random process into a numeric
value.
Definition
• Given a random experiment with sample space S, a
random variable X is a set function that assigns
numerical values to each element s of the sample
space.
• Example: Consider an experiment of tossing a fair coin
twice.
• The sample space for this experiment is
S = {HH, HT, TH, TT}.
• These outcomes are non-numerical.
• Define the variable X as:
X = the total number of heads obtained
• In tossing a coin twice, we may get two heads, one head
or no head depending on chance.
• Hence, X = the number of heads is a result of chance
and, thus, a random variable.
– X(HH) = 2
– X(HT) = 1
– X(TH) = 1
– X(TT) = 0
• A random variable can also be defined as a variable that
takes on different values with specified probabilities.
• In the above example:
– P(X = 2) = P(2 heads) = P(HH) = 1/4
– P(X = 1) = P(1 head) = P(HT) + P(TH)
= (1/4) + (1/4) = 1/2
– P(X = 0) = P(no head) = P(TT) = 1/4
Types of random variables
• If a random variable takes on a countable number of
values (often integers) such as 0, 1, 2, 3, …, then it is
called a discrete random variable.
• Example:
– The number of customers that arrive at a bank per
hour
– The monthly number of TV sets a store sells
• If a random variable takes on any numerical value
within a certain interval, then it is called a continuous
random variable.
• It is a random variable with a set of possible values that
is infinite and uncountable.
• For example, if we let X denote the daily revenue of a
randomly selected firm (in Birr), then X is a continuous
random variable since it can take on any real number.
Probability distributions
• The probability distribution of a random variable X is a
table, graph or mathematical formula that gives the
probability of:
– observing each value of X (if X is a discrete random
variable)
– X assuming any value in an interval (if X is a
continuous random variable)
• Example: Consider an experiment of tossing a fair coin
twice.
• Define the random variable X as: X = the number of
heads obtained.
• In two tosses of a coin, the possible outcomes are HH,
TH, HT and TT.
• Thus, the total number of heads we can get is 2, 1 or 0.
• The probability that the random variable X can take on
some specific value x is written as:
P(X = x), x = 0, 1, 2
• The probability that two tosses of a coin result in two
heads (x = 2) is:
P(X = 2) = P(HH) = 1/4
• The probability of obtaining one head (x = 1) is:
P(X = 1) = P(TH) + P(HT)
= (1/4) + (1/4) = 1/2
• The probability of obtaining no head (x = 0) is:
P(X = 0) = P(TT) = 1/4
Discrete and continuous probability distributions
• Probability distributions are either continuous or discrete
depending on whether they define probabilities for
continuous or discrete random variables.
• A continuous probability distribution describes the
probabilities of the possible range of values of a
continuous random variable.
• A discrete probability distribution describes the
probability of occurrence of each value of a discrete
random variable in its domain (R).
• Condition (i) says that the probability of any value
assumed by the random variable X should lie between 0
and 1 (inclusive).
• Condition (ii) tells us that the sum of the probabilities
of all the values X can take on is equal to one.
• One consequence of properties (i) and (ii) is that:
𝟎 ≤𝑷 𝒙 ≤𝟏
The normal distribution
