Sie sind auf Seite 1von 18

A probability is a numerical way of describing how likely (or not) an event is to happen.

If each of the elements in the sample space are equally likely, then we can define the
probability of event A as:

Example 1.
Find the probability of rolling an even number on an ordinary dice.

We have a sample space of S ={1, 2, 3, 4, 5, 6}.
Defining “throwing an even number” as event A, we have A = {2, 4, 6}.
So the probability of throwing an even number is given by:

Example 2.
One card is picked from an ordinary pack of 52 playing cards. What is the probability of
(i) a diamond
(ii) an ace
(iii) the ace of diamonds
(iv) a jack, queen or king.


Basic probability axioms

The three basic probability axioms can be summarised as follows:

1. P(S) =1
The relative frequency of an event that is certain to occur must be 1. The sample
space, S contains all possible outcomes and therefore the probability of S must be 1.
For example, when we are rolling a dice S ={1, 2, 3, 4, 5, 6}. When A=S, ie the event of
getting any one of the numbers 1 to 6 on a dice, the probability is:
The probability of getting any one of the numbers 1 to 6 on a dice is certain! So this result is
saying that we give certain events a probability of 1.

It follows that for event A from sample space S that, P(A¢)=1-P(A).

Suppose the probability of an individual dying is 0.2, then the probability of the individual
not dying is 1-0.2=0.8.

The relative frequency of occurrence of any event must not be negative, that is,
probabilities can never be negative.
If we define event A as “rolling a 7 on a dice”, then the probability of this is:

So the probability of an impossible event is 0.

So rules 1 and 2 together are telling us that probabilities lie between 0 (impossible) and 1

If two events cannot occur simultaneously, because they are mutually exclusive, the
probability of an event defined by their union is equal to the sum of the probabilities
of the two events. This property is known as additivity.

The reason this rule works is that there is no ‘overlap’ between the elements in A and B.
The rule has broken down because the element 1 is in both events A and B. This element has
been counted twice on the RHS of the equation, once in P(A) and once in P(B).

Example 3.
One card is picked from an ordinary pack of 52 playing cards. Consider the following events
A = {pick a 7}, B = {pick an ace} and C = {pick a club}.
Show that P(A UB) = P(A) + P(B):
(i) works for events A and B
(ii) doesn’t work for events A and C and explain why.


This rule works is that there is no ‘overlap’ between the elements in A and B.

This is because the 7 of clubs is in both A and C and so is counted twice on the RHS of the

The addition rule

Axiom 3 is often known as the special addition rule. For a more general case, where
two sets are not necessarily mutually exclusive, the rule can be extended as follows:
Example 4.
A contestant on a game show is asked two questions. The probability that she gets the first
question correct is 0.3 and the probability that she gets the second question correct is 0.4.
Given that the probability that she gets both questions correct is 0.1, calculate the probability
(i) she gets either the first, the second or both questions right
(ii) she gets both questions wrong.

If A = {get 1st question correct} and B = {get 2nd question correct} then we are told:

(ii) We want the probability that both questions are wrong, that is the complement of
getting at least one question right. Hence the probability is:

Conditional probabilities
Consider the two events, A and B. We might wish to know the probability that event A
occurred, given the occurrence of the event B. This is known as a conditional
probability and is denoted thus:
P(A|B) is read as “the probability of event A occurring given that event B has already
occurred” or “probability of A given B” for short. This is called a conditional probability as
the probability depends (ie is conditional) on event B.
The conditional probability of A occurring given B can be expressed as:

The above formula can be explained as representing the occasions that event A
occurs with B relative to the occasions that B occurs (with or without A).

Example 5.
Consider picking a card from an ordinary pack of playing cards. If we have the events:
A = {pick a spade} B = {pick an 8}
calculate the probability of picking a spade given that we have picked an 8, ie calculate

Since there are only four 8’s in the pack and only one of them is a spade, we conclude that

Hence, solution can be rearranged as:

The multiplication rule

Rearranging our formula for conditional probabilities, we get:
Independent events
Events A and B are said to be independent if whether or not event B has occurred
gives us no information on whether event A has occurred. This can be expressed
algebraically as follows:

This is a special case of the multiplication rule when events A and B are independent.

Example 6.
Two dice are thrown. Find the probability of rolling a 5 on both dice.

Since these events are independent:

Example 7.
A card is picked from an ordinary pack of 52 playing cards, without replacement, and
then another one is picked. What is the probability of picking:
(i) two red cards
(ii) one of each colour?

A tree diagram is used to show the possible outcomes. In this case we can either get red or a
black card. So the number of branches is the number of different possibilities, and the
numbers on the branches are the probabilities of those particular events happening.
Notice that the probabilities on any particular set of branches add up to one.
To calculate probabilities, go along the branches of the tree from left to right to get to the end,
and multiply together any probabilities that you have passed – using the multiplication rule.
If there is more than one route through the tree to give the answer you require, then sum the
answers from the different routes – using the addition rule for mutually exclusive events.

Example 8.
A box of chocolates contains 8 milk chocolates and 4 plain chocolates. A chocoholic eats
three chocolates. Calculate the probability that:
(i) all three are milk chocolates
(ii) exactly one is a plain chocolate.

If M is the event “eats a milk chocolate” and P is the event “eats a plain chocolate”, then the
tree diagram is as follows:
We could also have solved this more quickly using combinations:

Example 9.
In a restaurant, 45% of the customers are female. 74% of females choose from the à la carte
menu, whilst only 37% of males do. The rest choose from the set menu. What is the
probability that:
(i) a customer orders from the set menu
(ii) a customer ordering from the à la carte menu is female?

If F is the event “is female”, M is the event “is male”, A is the event “chooses from à la
carte”, and S is the event “chooses from the set menu”, then the tree diagram is as follows:
Law of total probability
Consider the set space, S , being divided into a partition of n mutually exclusive
events, Ei where i= 1, 2, 3,… ,n, then:

For example, a partition of our dice sample space S ={1, 2, 3, 4, 5, 6} could be:

There are many other possibilities, as long as the Ei ’s are mutually exclusive (ie there is no
overlap between them) and together make up the whole sample space (ie they are exhaustive)
it is a partition.

For example, if A is the event “roll an even number” we have A = {2, 4, 6}. This can be
made up of all the intersections with our partition from above:


This result is known as the law of total probability.

Bayes Theorem
This theorem is named after the Reverend T Bayes and is used extensively in
Bayesian methods of statistical inference.
The result is as follows:

Using the conditional probability:


also, the relationship:

and the law of total probability:

The result is:

Essentially Bayes formula allows us to ‘turnaround’ conditional probabilities, ie calculate
P(Ei |A) given only information about P(A|Ei ) .
The values P(Ej) are known as prior probabilities, the event A is some event which is
known to have occurred and the conditional probability P(Ei|A) is known as the
posterior probability.

Example 10.
A person has three routes to get to work. The probability that he arrives on time using routes
A, B and C are 60% , 62% and 70% respectively. If he is equally likely to choose any of the
routes, and arrives at work on time, what is the probability that he chose route B?

If A is the event “chooses route A”, B is the event “chooses route B”, C is the event “chooses
route C”, and T is the event “arrives on time to work”, then: