Chapter 2

Probability and counting problems

MATH1108: Probability & Statistics 1 University of Nottingham 1 / 22

2.1 Probability
Definition
A probability measure is a real-valued set function P defined on the
events of a sample space Ω which satisfies the following three axioms:

A1. P(E ) ≥ 0 for any event E .

A2. P(Ω) = 1.
A3. If E1 , E2 , . . . , En are disjoint (i.e. Ei ∩ Ej = ∅ for i ̸= j) then
P(E1 ∪ E2 ∪ . . . ∪ En ) = P(E1 ) + P(E2 ) + . . . + P(En ).

If Ω is infinite (A3) is replaced by

A3’. If E1 , E2 , . . . are disjoint (i.e. Ei ∩ Ej = ∅ for i ̸= j) then
P(E1 ∪ E2 ∪ . . .) = P(E1 ) + P(E2 ) + . . ..

Note that Axioms A1 and A3 (A3′ ) are common to all measures. We

are all used to measures, e.g. time, length and weight.
MATH1108: Probability & Statistics 1 University of Nottingham 2 / 22
Probability

Some basic results

(i) P(E c ) = 1 − P(E ).

Proof:

1 = P(Ω) (A2)
= P(E ∪ E c )
= P(E ) + P(E c ), (A3)

so P(E c ) = 1 − P(E ).

MATH1108: Probability & Statistics 1 University of Nottingham 3 / 22

Probability

(ii) P(∅) = 0.

Proof:
Set E = ∅ in (i) and recall that ∅c = Ω, to give
1 − P(∅) = P(Ω) = 1 by (A2),
hence P(∅) = 0.

(iii) P(E ) ≤ 1.

Proof: From the proof of (i),

1 = P(E ) + P(E c ).
But P(E c ) ≥ 0 (A1), so P(E ) ≤ 1.

MATH1108: Probability & Statistics 1 University of Nottingham 4 / 22

Probability
(iv) P(E ∪ F ) = P(E ) + P(F ) − P(E ∩ F ).
Proof:
Write both E ∪ F and F as disjoint unions:

E ∪ F = E ∪ (F ∩ E c )
F = (E ∩ F ) ∪ (F ∩ E c )

Now apply (A3) to the above:

P(E ∪ F ) = P(E ) + P(F ∩ E c )

P(F ) = P(E ∩ F ) + P(F ∩ E c )

Subtracting,
P(E ∪ F ) − P(F ) = P(E ) − P(E ∩ F )
and therefore
P(E ∪ F ) = P(E ) + P(F ) − P(E ∩ F ).
MATH1108: Probability & Statistics 1 University of Nottingham 5 / 22
Probability
(v) P(E ∪ F ) ≤ P(E ) + P(F ).
Proof:
By (A1), P(E ∩ F ) ≥ 0. Thus, from (iv)

P(E ∪ F ) = P(E ) + P(F ) − P(E ∩ F )

≤ P(E ) + P(F ).

(vi)

P(E ∪ F ∪ G ) = P(E ) + P(F ) + P(G ) − P(E ∩ F ) − P(E ∩ G )

− P(F ∩ G ) + P(E ∩ F ∩ G ).

Proof:
See Exercise Sheet 1.
MATH1108: Probability & Statistics 1 University of Nottingham 6 / 22
Probability

Examples

1. In many cases, Ω consists of N equally likely elements, ie

1
Ω = {ω1 , ω2 , . . . , ωN } , P(ωi ) = .
N
Then, for any event E (ie subset of Ω),

#(E )
P(E ) =
N
where #(E ) denotes the number of elements in E .

MATH1108: Probability & Statistics 1 University of Nottingham 7 / 22

Probability

Example:
1.Throw a die; Ω = {1, 2, 3, 4, 5, 6}.

The probability of each outcome is 1/6 :

1
P({1}) = P({2}) = . . . = P({6}) =
6
The probability that the roll of a die results into an odd number can
be calculated as follows:
3 1
P(Odd) = P({1, 3, 5}) = =
6 2

MATH1108: Probability & Statistics 1 University of Nottingham 8 / 22

Probability
Example:
2. Draw a card at random from a standard pack of 52.

Sample Space: Ω = {A♣, 2♣, 3♣, . . . , K ♢}.

Probability of drawing a specific card: P(ω) = 1/52 for all ω ∈ Ω.

If E = {Black}, F = {King}, then

P(E ∪ F ) = P(E ) + P(F ) − P(E ∩ F )

26 4 2
= + −
52 52 52
7
= .
13

MATH1108: Probability & Statistics 1 University of Nottingham 9 / 22

Probability
Events are defined in terms of sets (collections of elementary events).
For sets E and F we talk in terms of:
union (E ∪ F ) - the elements which belong to at least one of
two sets;
intersection (E ∩ F ) - the elements which belong to both of
two sets;
negation (E /F ) - the elements which belong to one set (E ) but
not to the other set (F ). Note that E /F = E ∩ F c .

However, sets do not in themselves have numerical value, so writing

E + F does not make any sense and is wrong. Please note that
E − F is sometimes used for negation E − F = E /F and E × F will
be defined elsewhere.

MATH1108: Probability & Statistics 1 University of Nottingham 10 / 22

Probability

Probabilities are special examples of measures. Measures assign

(numerical) values to sets. For example, the simplest measure is the
otal number (count) of objects in a set. Since measures take
numerical values, we can add, subtract and multiply the probabilities
assigned to sets E and F . i.e.

P(E ) + P(F ), P(E ) − P(F ), P(E ) × P(F )

However, probabilities are not sets, so writing

P(E ) ∪ P(F ), P(E ) ∩ P(F )

does not make sense and is wrong.

MATH1108: Probability & Statistics 1 University of Nottingham 11 / 22

2.2 Counting Problems
Multiplication Rule

The principal axioms of counting are:

1. The number of ways of selecting one object from n objects is n.

2. The number of ways of selecting one object from n objects and
another one from m objects is n × m.

This leads to the following: If an operation has k distinct stages and

stage r has Nr possible outcomes, then the total number of outcomes
of the operation is

N 1 × N2 × · · · × Nk

MATH1108: Probability & Statistics 1 University of Nottingham 12 / 22

2.2 Counting Problems

Example:

Suppose we flip three fair coins and roll two fair six-sided dice. Each
coin has two possible outcomes (heads and tails), and each die has
six possible outcomes {1, 2, 3, 4, 5, 6}. The total number of possible
outcomes of the three coins and two dice is thus given by multiplying
three 2’s and two 6’s, i.e.,

2 × 2 × 2 × 6 × 6 = 288

MATH1108: Probability & Statistics 1 University of Nottingham 13 / 22

Counting Problems
Example:
A student hall has 40 Science students, 30 Arts students and 20
Medical students. A committee of 3, consisting of 1 student from
each faculty, is to be chosen. Then the number of ways to form the
committee is
40 × 30 × 20 = 24000.

Example:
How many different ways of putting 4 distinct letters into 4 distinct
envelopes?
First fill the first envelope - 4 letters to choose from. For the second
envelope there are 3 letters to choose from, then 2 and 1 for the
third and fourth envelopes respectively.
So total number of ways is 4 × 3 × 2 × 1 = 4! = 24.
MATH1108: Probability & Statistics 1 University of Nottingham 14 / 22
Counting Problems

Suppose four friends go to the theatre and each leaves their coat in
the cloakroom. At the end of the play, the four coats are randomly
returned to the four friends. Here the total number of different ways
the coats can be returned is equal to 4 × 3 × 2 × 1.

Definition
If r is a positive integer then r -factorial written r ! is

r ! = r × (r − 1) × · · · × 3 × 2 × 1
with the convention 0! = 1.

MATH1108: Probability & Statistics 1 University of Nottingham 15 / 22

Counting Problems

Definition
A permutation of length r from n distinct objects is an ordered
arrangement of r of the objects.
The number of permutations of length r of n objects is

n n!
Pr = n(n − 1)(n − 2) . . . (n − r + 1) =
(n − r )!

Note that n Pn = n!.

MATH1108: Probability & Statistics 1 University of Nottingham 16 / 22

Counting Problems
Example: Raffle tickets numbered 1, 2, . . . , 500; 3 tickets drawn,
without replacement, one at a time for 1st, 2nd and 3rd prizes.
Ω = {(a, b, c) : 1 ≤ a, b, c ≤ 500, a ̸= b ̸= c ̸= a}.#(Ω) =500 P3 = 500 × 499 × 498.

E = {ticket number 1 wins 1st prize, ticket number 2 wins 2nd prize}.
#(E ) = 498, as 1st and 2nd prizes are fixed.

498 1
Thus P(E ) = 500×499×498
= 500×499
.

F = {ticket number 1 wins a prize} = F1 ∪ F2 ∪ F3 ,

where Fi = {ticket number 1 wins ith prize}.

F1 , F2 and F3 are disjoint =⇒ P(F ) = P(F1 ) + P(F2 ) + P(F3 ).

499 × 498 3
#(F1 ) = 499 × 498 = #(F2 ) = #(F3 ) =⇒ P(F ) = 3 = .
500 × 499 × 498 500

MATH1108: Probability & Statistics 1 University of Nottingham 17 / 22

Counting Problems

Definition
A combination of r objects from a collection of n distinct objects is
an unordered arrangement of r of the objects. The number of such
combinations is
  n
n n Pr n!
Cr = = = .
r r! (n − r )!r !

Note that nr = n−r n

.

(Proof 1: Use definition above.

Proof 2: It’s obvious - think about it.)

MATH1108: Probability & Statistics 1 University of Nottingham 18 / 22

Counting Problems
Example:
15 people, 6 men and 9 women. Committee of 5 chosen at random,
i.e. all combinations equally likely.
Ω = {All possible combinations of 5 from 15}, so
 
15
|Ω| = #(Ω) = = 3003.
5

Define E as the event that the committee has 3 men and 2 women.
Then,    
6 9
#(E ) = × = 20 × 36 = 720.
3 2
Thus
6

9

#(E ) 3 240
P(E ) = =
2 =
15 .
#(Ω) 5
1001

MATH1108: Probability & Statistics 1 University of Nottingham 19 / 22

Counting Problems

Example

Pick 5 cards from a pack at random. What is the probability of a

straight? (Assume Ace may be high or low.)

Straight = 5 consecutive cards, not all of same suit (eg 4H, 5C, 6D,
7D, 8H.)

Ω = {All unordered 5-tuples}, so #(Ω) = 52

5
.
E = {All straights}.

MATH1108: Probability & Statistics 1 University of Nottingham 20 / 22

Counting Problems
To find #(E ):
Fix a starting card (must be one of 1(A), 2, . . . , 10, ie 10
possibilities.) Then each card in the run can be one of 4 suits, so
(multiplication rule) total number of suit choices for the 5 cards is 45 .

But cards cannot all be same suit, so subtract 4 choices (all H, all C,
all D or all S). Thus

#(E ) = 10 × (45 − 4).

Thus,
#(E ) 10(45 − 4)
P(E ) = = 52

≈ 0.0039.
#(Ω) 5

MATH1108: Probability & Statistics 1 University of Nottingham 21 / 22

Exercises

Exercise sheet 1: 5-9

MATH1108: Probability & Statistics 1 University of Nottingham 22 / 22