Chapter 2 ELEMENTARY SET THEORY

We adopt the naive (as opposed to axiomatic) point of view for set theory and regard the notions of a set as primitive and well-understood without formal denitions. We just assume that a set A is a collection of objects characterized by some dening property that allows us to think of the objects as a whole entity. The dening property has to be such that it must be clear whether a given object belongs to the set or not. The objects possessing the property are called elements or members of the set. We denote sets by common capital letters A, B, C, etc. and elements or objects of the sets by lower case letters a, b, c,etc. For example, we write A = {a, b, c}
to indicate that and if

2.1

Introduction

2.1.1 Description of a Set

is a collection of elements

belongs to

we write

x A.

If

a, b, c.

If

is any element, if

doesnt belong to

we write

A is a x / A.

set,

To describe a particular set, we have to indicate the property that characterizes its elements. To start, lets consider a set which has no members. Since a set is determined by its elements, there is only one such set which is called the

set, which is denoted by . Any be non-empty or non-void. Denition 34

set

A,

empty set,

or the

null

consisting of one or more elements is said to

A non-void set A is said to be nite if A contains n distinct elements, where n is some positive integer. Such a set A is said to be of order n. The null set is dened to be nite with order zero. A set consisting of exactly one element, say A = {a}, is called a singleton . If a set A is not nite, then we say that A is
innite.

To describe a nite set, one way to proceed is by listing or enumerating

members. For example, integers 1 3 and 5.

S = {1, 3, 5} means that S

its

is the set whose elements are the

Note that we use braces to denote the set and that the order

of the elements plays no role in describing the set. Thus the sets =

{5, 3, 1}

{1, 3, 5}

and

are to be viewed as exactly the same set.

20

Elementary Set Theory

But in general, and especially for innite sets , we do not describe a set by listing all its members. A convenient method of characterizing sets is as follows. Suppose that for each elements x of a set A there is a propositional function P (x) which is either true or false. We may then dene a set C which consists of all elements x A such that P (x) is true, and we may write
C

= {x A : P (x) is true}.

For brevity of notation, we will omit the words is true and, when it is clear which set x belongs to, we will also omit the reference set A. Thus we will sometimes write {x : P (x)} . From a given set, we can form new sets, called subsets of the given set. In
general we say that a set A is a subset of a set belongs also to B. More formally

whenever every element of

Denition 35 Given two sets A and B, we say that A is a subset of B , or that A is contained in B, and denote it A B (or B A), i x : x A x B. We say that A is a proper subset of B, or that A is strictly contained in B , and denote it A B, i [A B (B A)] is true. Denition 36 We say that a set A is equal to B and write A = B i x A x B, A B A = B [A B (B A)] A dierent B A=B A B x y x = y. x y x = y. Exercise 37 Show that if A is any set, then A A and A. Exercise 38 Show that A = B [A B (B A)] . Denition 39 Let X be a set and let A X.The complement of subset A with respect to X is the set of elements of X which do not belong to A. We denote the complement of A with respect to X by AC / A}. X = {x X : x
i.e., and have the same elements. Equivalently is from . We say that , and we write , if and are not equal. If write and If denote the same element of a set we say that they are equal and we denote distinct elements of a set, we write and

When it is clear that the complement is with respect to X , we simply say the and write AC = {x X : x / A}. In every discussion involving sets, we will always have a given xed set in mind from which we take elements and subsets. We will call this set the universal set, and we will usually denote it by X.
complement of A,

Throughout the remainder of the present chapter, X will denote always an arbitrary non-empty xed set.

Exercise 40 Prove the following statements. Let A,B, C be subsets of X. Then

1. if A B and B C then A C.

Operation on sets

21

2. XC = 3. C = X 4. AC C = A 5. A B B C AC 6. A = B AC = BC

a nonempty set whose only element is the empty set, i.e. {}. In this case, {} is a singleton. We see that {} and {}. There is a special family of subsets of to which we have given a special name.

X or a collection of subsets of X. In terms of notation, this concludes our hierarchical system, where lowercase letters refer to elements of X, upper case letter to subsets of X and script letters to families of subsets of X. Lets note that the empty set is a subset of X. It is possible then to form
family of subsets of

Next we need to consider sets whose elements are sets themselves. For example, if A,B and C are subsets of X , then the collection A = {A,B,C } is a set, whose elements are A,B and C. We usually call a set whose elements are subsets of X a

Denition 41

by

power set

P (A) .

Let A be any subset in X. We dene the power class of or the of to be the family of all subsets of . We denote the power set of A Specically

P (A) = {B : B A} The power class of the empty set is P () = {},i.e. the singleton of . The power class of a singleton is P ({a}) = {,{a}}.Notice that the power set of A always contains A and . In general, if A is a nite set with n elements, then P (A) contains 2n elements. This is why sometimes the power set of A is (improperly) denoted as 2A.

Exercise 42 Prove that if A is a nite set with n elements, then P (A) contains 2n
elements.

2.2 Operation on sets

2.2.1 Union and Intersection Denition 43 Let A and B be two subsets of X.We dene the union of sets A and B , denoted by A B, as the set of all elements that are in A or B; i.e.

A B = {x X : [(x A)

(x B )]}.

22
intersection

Elementary Set Theory

We also dene the of sets A and B , denoted by A B, as the set of all elements that are in both A and B; i.e.
Both denitions can be extended to nite or innite collection of sets. Denition 44 Two sets are said to be disjoint if they dont have any element in common, i. e.. if A B = .

A B = {x X : [(x A) (x B)]}.

Exercise 45 Prove the following properties 1. A AC = ; A AC = X.

2. Commutative laws:

A B = B A; A B = B A. A (B C ) = (A B) C = A B C ; A (B C ) = (A B) C = A B C.

3. Associative law:

4. Distributive law:

A (B C ) = (A B) (A C ); A (B C ) = (A B) (A C ). 5. (A B ) [(A B ) = B ] [(A B ) = A]

6. De Morgans laws

(A B )C = AC B C (A B )C = AC B C

From the associative law established in the previous exercise, there is no ambiguity in writing A B C. Extending this concept, let n be any positive integer and let A1 , A2 ..., An denote subsets of X. The set A1 A2 ... An is dened to be the set of all x X which belong to at least one of the subsets Ai , and we write
n
i

=1

A = A1 A2 ... An = {x X : x Ai for some i = 1, ...,n}

i

Using a similar argument we dene the intersection of

n
i

n subsets of X as

=1

A = A1 A2 ... An = {x X : x Ai for all i = 1, ...,n}

i

Relations

23

Theorem 46

[

Let A1 , A2 ..., An
] =

be subsets of

X. Then

Proof. Is left as an exercise. Use your results from the previous exercise.
Relative Dierence

Ai C n i=1 A
n i=1

n i=1
n i

C = i]

AC i =1 A
C i

2.2.2

B A) , also known as the relative complement of A in B is dened as the set of elements in B that do not belong to A. Hence B A = {x X : xB x / A}.
A , denoted as (

Denition 47 Let A and B be two subsets of X. The relative dierence of B and

Exercise 48 Prove the following statements. Let A and B be two subsets of X. Then
1. A B = A B C .
2. 3. 4. 5.
(

A B )C = AC B. A = B A B = B A. B (B A) = A A B. BA=B BA=
n i i

Exercise 49 Let B and A1, A2..., An be subsets of X. Then

1. B [ n i=1 A ] =
2.

B[

n i=1

Ai

=1 (B Ai )

] =

n i=1 (

BA)
i

2.3 Relations

So far the order of the elements has not been relevant. Thus the set {a, b} has been considered equivalent to the set {b, a} . There are times however where the order is important. When we wish to indicate that a set of two elements a and b is ordered, we enclose the elements in parentheses: (a, b) . Then a is called the rst element and b the second element. A set of such two elements is called an ordered pair, and is dened in the following way.

2.3.1 Ordered pairs

Denition 50
That is

(a, b) = {{a}, {a, b}}.

An ordered pair (a, b)

is a set whose members are

{a}

and

{a, b}.

From the denition, we can derive a main property of ordered pairs.

24

Elementary Set Theory

Theorem 51 (x, y) = (a, b) x = a and y = b Proof. Lets rst prove the converse. If x = a and y = b then (a, b) = {{a} , {a, b}} = {{x} , {x, y}} = (x, y) Now lets assume (x, y ) = (a, b) . Then by denition we have {{x} , {x, y }} = {{a} , {a, b}} . Consider two cases, depending on whether x = y or x = y. If x = y then {x} = {x, y} ,so (x, y) = {{x}} . Since (x, y) = (a, b) then we have {{x}} = {{a} , {a, b}} .
The set on the left hand side has only one member x .Thus the set on the right must also have one member, so a = a, b , concluding that a = b. Then x = a , so x = a and x = a. Thus x = y = a = b. If x = y, then it must be a = b otherwise we would be arriving to a contradiction . From the assumption we know that (x, y) = (a, b) , so

{} {}

{} { }

{}

{{ }} {{ }}

a = x. By the same argument we have

{x} {{a} , {a, b}} implying that {x} = {a} or {x} = {a, b} . In either case we have a {x} , and so {x, y} {{a} , {a, b}} and given that x = y it has to be {x, y } = {a, b} . Now x = a, x = y and y {a, b}
implies y

= b.

Exercise 52 Extend the denition of an ordered pair to a nite set of n elements.

Call it n-dimensional vector or n-tuple. 2.3.2 Cartesian Product Denition 53 The Cartesian product of two sets A and B, written A B, is the set of all ordered pairs (a, b) such that a A and b B. That is

Notation We will write A2 = A A.

2.3.3 Properties
1. (

AB

= {(a, b) : a A,

and

b B}

It is possible to derive some important properties, which we enumerate below.

A B ) (C D) = (A C ) (B D). 2. (A B ) C = (A C ) (B C ). Exercise 54 Prove the two properties stated above.

Relations

25

2.3.4 Binary Relations

is either true or false. For example, the relation less than is a relation between positive integers. When considering a relation between two objects, in some cases it is necessary to know which object comes rst. For instance, if Mark is taller than Paul is true then Paul is taller than Mark is false. Thus, it is natural for the formal denition of a relation to depend on the concept of an ordered pair.

A binary relation between two objects a and b is a condition involving a and b that

Denition 55 Let A and B be sets. A relation between A and B

R of A

B. We say that

denote this by saying a

relation on A.

a A and b B are related by R if (a, b) R, and we aRb. If B = A, then we speak of a relation R A A being

is any subset

Denition 56 Let R be a relation between two sets A and B. The domain of R,

R, denoted R(R), is the set of all second elements of R. Formally, R(R) = {b B : a A (a,b) R}.Clearly from the denition we obtain D (R) A, and R(R ) B.
(
The

D(R) {a A : b B
denoted

, is the set of all rst elements of members of R. Formally,

a, b) R}.

range

D(R) =

of

Denition 57 A binary relation R on a set A can have the following properties: Completeness: a, b A, aRb bRa. Reexive property: a A, aRa. Irreexive property: a A, (aRa).
Symmetric property: a, b A, aRb bRa. Antisymmetric property: a, b A and a = b, aRb (bRa) . Transitive property: a, b, c A, (aRb bRc) aRc.

Example 58

1. The relation dened on S = {1, 2, 3, 4, 5} is complete, reexive and antisymmetric.

2. Let S be the set of all people who live in the US, and suppose that two people x and y are related by R if x lives within a mile of y. Then R is not complete, is reexive and symmetric, but not transitive.

Exercise 59 Let R be the relation to be taller and younger than on the set S of all
the Econ 897 students.
What properties does R have?

We are specically interested in two types of relations which come up frequently in economics and mathematics, namely equivalence and order relations.

26

Elementary Set Theory

2.3.5

Equivalence relations A relation R on a set S is an equivalence relation if it is reexive, symmetric and transitive. An equivalence relation is often denoted by = or .

Denition 60

Note that completeness is not a requirement for equivalence. This type of relation is singled out because it possess the properties naturally associated with the idea of equality.

Example 61 Let S be the set of all people who live in the US, and suppose that two
people x and y are related by R if x has the same age as y, then R is reexive, transitive and symmetric. Hence R is an equivalence relation.

Exercise 62 Let S be the set of all lines in a plane. Determine which of the following
are equivalence relations x R y if x is parallel to y x R y if x is perpendicular to y

Given an equivalence relation R on a set S, it is natural to group together all the elements that are related to a particular element. More precisely
an element of

Denition 63 Let R
A.

{y A : yRx} .

The

be an equivalence relation on a set A, and let x be equivalence class of x with respect to R is the set Ex =

A2

Exercise 64 Prove the following statements:

1) Let S be an equivalence relation dened on a set A. Then every two dierent equivalence classes must be disjoint. 2) Let S be an equivalence relation dened on a set A. Then every element of A belongs to at most one equivalence class.

Thus we see that an equivalence relation R on a set S breaks S into disjoint subsets in a natural way. This operation on a set is known as a partition.
2.3.6 Partition Denition 65 Let A be a subset of X. A partition of A is a collection
nonempty subsets of 1. Each 2.

A such that

of

x A belongs to some subset S A. For all S, T A, if S = T, then S T = .

Relations

27

Example 66 Let S be the set of all students in a particular university. For x and
y in S, dene xRy i x and y were born in the same calendar year. Then R is an equivalence relation, and a typical equivalence class is the set of all students who were born in the same year as student z.

Exercise 67 Consider the previous example. If y Ez , does this mean that y and z

are the same age?. Would it have been the same to dene the equivalence relation as the set of all students who were born in a particular year, say in 1980?

Not only does an equivalence relation on a set S determine a partition of S, but the partition can be used to determine the relation. We formalize this as a theorem

Theorem 68

Let R be an equivalence relation on a set S. Then A = {Ex : x S } is a partition of S . The relation belongs to the same piece as is the same as R. Conversely, if P is a partition of S, let R be dened by xRy i x and y are in the same piece of the partition. Then R is an equivalence relation and the corresponding partition into equivalence classes is the same as P .

class classes must be either equal or disjoint. Hence all the dierent equivalence classes have empty intersection, making A a partition on S . Now suppose that P is the relation belongs to the same piece (equivalence class) as . Then

Proof. Since R is reexive, each element x S belongs to some equivalence Ex A. Moreover, from a previous exercise, we know that two equivalence

xPy x, y Ez for some z S xR zand yRz for some z S xRy

Thus P and R are the same. Conversely, suppose that P is a partition of S and let R be dened by xRy i x and y are in the same piece of the partition. Clearly, R is reexive and symmetric. To see that R is transitive, suppose that xRy and yRz . We know that y A for some A P , therefore x A , z A and hence xRz. Finally, the equivalence classes

2.3.7 Order relations Denition 69 A binary relation is called a quasi-order or preorder if it is reexive and transitive.

of R correspond to the pieces of

because of the way

was dened.

Example 70

Let S be the set of all students in a particular university. Let R be the relation is at least as tall as and as old as. Then R is not complete, but is reexive and transitive. Then it is a preorder.

28

Elementary Set Theory

Denition 71 A binary relation is called an order if it is, reexive, transitive, and antisymmetric. It is called a strict order if it is irreexive, transitive and antisymmetric. Exercise 72 Is the preorder of the previous example an order? Justify. Denition 73 An order that is also complete is called a complete or total order. Notation To stress that some orders may not be complete, we sometimes call them
a < b indicates a strict y (>)x instead of x

partial orders. An order is often denoted by or . The notation

order where a (<)y.

and a = b . It is equivalent to write

An example of partial order is set inclusion; for each set X, the relation is a partial order in the power set P (X ). The reason for qualifying partial is that sometimes the order between two elements is not dened, i.e., there can be M, N P (X ) such that neither M N nor N M.
Denition 74

An ordered set is a set S in which an order is dened.

the statements x < y, x = y, y < x is true.

Again, the order can be either partial or complete. For the case of a complete ordered set ,the following statement can be shown Exercise 75 Let S be a complete ordered set. Then x, y S, one and only one of

2.4 The Set of Natural Numbers. Induction

Lets assume the existence of a set N
natural numbers. Thus consisting of the positive integers, also called

N = {1, 2, 3, 4...}
This set can be characterized from the following set of Axioms, called the Peano axioms in honor of the Italian mathematician Giuseppe Peano, who developed this approach in the late 19th century. We suppose that there exist a set P whose elements are called natural numbers and a relation on P called successor with the following properties:

P1.

There exists a natural number, denoted by 1, that is not the successor of any other natural number.

Note that any complete relation is also reexive. This makes our denition of complete ordering somehow redundant, given that one requirement is implied by another. Nevertheless this way of dening an order is customary (at least in economics) and we decided to stick with the widely adopted denition.

The Set of Natural Numbers. Induction

29
denote

P2. Every natural number has a unique successor. If m P, then we let m

the successor of

m.

P3. Every natural number except 1 is the successor of exactly one natural number. P4. If M is a set of natural numbers such that
1. 1 M and 2. k P, if k M , then k M,then M = P.
Axioms P1 to P3 express the intuitive notion that 1 is the rst natural number and that we can progress through the natural numbers in succession one at a time. Axiom P4 is the equivalent of what we will call later the principle of mathematical induction.

Exercise 76 Consider the relation successor on N. What are its properties? Using the previous axioms, we can dene what addition and multiplication means. Basically

D1. n P, dene n + 1 = n . D2. Let n, m P. If m = k and n + k is dened, then dene n + m = n + k

(n + k ) .

D3. n P, dene n 1 = n. D4. Let n, m P. If m = k and n k is dened, then dene n m = n k

= (n k )+ n.

Note that in D2 and D4 the existence of k is assured by axiom P3. Based on the denition of sum, we can now introduce a strict order < and an order relations on N

Denition 77 Let N be the set of natural numbers. For x, y N

i n N : x + n = y. We also say that x y

i x < y Exercise 78 Verify that < is indeed a strict order on N

[(

x = y )]

we say that x < y

is true.

30

Elementary Set Theory

2.4.1 The Well Ordering Property and the Principle of Mathematical Induction There is an additional property of N that expresses in a precise way the idea that each nonempty subset of N must have a least (rst) element.

Theorem 79 (well-ordering property of N ) If S

there exists an element

Proof. Lets consider two cases: if 1 S, and if 1 / S. In the rst case, we know from axiom P1 that 1 is not the successor of any element of N. Then for every element k of S it must be 1 k. So m exists and is 1. If 1 / S the proof goes by contradiction. Suppose that there is no such least element m. We will construct the set M = {n N : n / S {1, 2, ..., n 1, n} S = } . This is the largest set of consecutive natural numbers starting at 1 that do not intersect with S. Clearly 1 M by assumption. Now if l M, then l = l + 1 must also belong to M. For if not then
l

mS

such that

mk

is a nonempty subset of

for all

k S.

N, then

and then it would be the least element of Therefore from axiom P4

S = , contradicting
S. the assumption that

implying that

= N,

is nonempty.

induction.
time.

One important tool to be used when proving theorems about the natural numbers comes from axiom P4 and is known as the principle of mathematical
It enables us to conclude that a given statement about natural numbers is true for all natural numbers without having to verify it for each number one at a

Theorem 80 (Principle of Mathematical Induction) Let P (n) be a statement that is either true or false for each n N. Then P (n) is true for all n N provided that

1. P(1) is true, and 2. For each k N, if P (k) is true then P (k + 1) is also true. Proof. The proof is trivial if we rely on axiom P4. Lets call S to the set
S = {k N : P (k )
is true

}
b)

k + 1) S . n.

From the assumptions we know that a) 1 From axiom P4 we have that

S and S = N, implying

that

N, k S P (n) holds for all

Exercise 81 Take the well ordering property as an axiom and prove the Principle
of Mathematical Induction without relying on axiom P4. (Hint: the proof goes by contradiction. Construct an auxiliary set S = {n N : P (n) is false} and use the
well ordering property to guarantee the existence of a least element

m S.)

for induction and part 2 as the induction step.The assumption that P (k ) is true in verifying part 2 is known as the induction hypothesis. It is essential that both
parts be veried to have a valid proof using mathematical induction.

It is customary to refer to the verication of part 1 of the theorem as the

basis

The Set of Natural Numbers. Induction

31

Exercise 82 Prove the following theorem: Let

that is either true or false for each n m. Then that 1. P m is true and 2. For each k ,if P k is true then P

m N and let P (n) be a statement

P (n) is true for all n m provided

( )

()

(k + 1) is true

their use, check the appendix of this chapter .

An useful and important theorem is proven by induction. Write x! for x(x 1)(x n 2) 3 2 1 for x 1, set 0! = 1 , and write n n. x for the ratio x nx for all x n n n+1 Observe that k + k1 = k for further details on combinatory numbers and
! !( )!

2.4.2

Applications

the formula of the binomial theorem holds for n (inductive step). Show then that it also holds for n + 1.

n n k k Theorem 83 The Binomial Theorem (a + b)n = n b . k=0 k a Proof. The formula is trivially valid for n = 1. (Basis of induction) Suppose
:

(a + b)

n+1

= (a + b)(a + b) = (a + b)
n

= = = = =

n k k=0 n n k k=0 n n+1

n+1

n n k k a b k k=0 n n nk k+1 ank+1bk + a b k k =0 n+1 n ank+1bk + an +1b l1 l=1

l l

by assumption

chg of var l=k+1

l

n+1

l=1

n l

n+1

n l=1

n + l an +1 b + 1 n + 1 n +1 n+1 a b + l n+1
l l l

n n+1 b n bn+1

n + 1 (n+1)k k a b k k=0

which is indeed the corresponding expression for n + 1.

Exercise 84 Prove or give a counterexample of the following statements 1. For every positive integer n, n2 + 3n + 8 is an even number. (n+1) n 2. n N, n i=1 i = 2 2 = (2n+1) (n+1) n 3. n N, n i i=1 6 3 = (n+1)2 n2 = ( n i)2 4. n N, n i i=1 i=1 4 xn+1 5. Assuming x = 1 , 1 + x + x2 + ... + xn = 1 1x , n N

32

Elementary Set Theory

6. n N , n > 3 n2 n ! 7. n N, n > 3 n2 2n 8.n N, n > 3 2n n!

2.5 Appendix: Binomial Coecients

Binomial coecients are used in many sciences for counting events and computing probabilities, and they constitute a useful mathematical tool in dierent areas in Economics.
1) (n 2)....2 1. The factorial of 0 is dened to be 0! = 1.

Denition 85 The factorial of a natural number n, n!, is computed as n! = n (n

Factorial numbers are useful for computing permutations of distinguishable elements. For example, suppose that you have to compute the number of possible

arrangements of 10 individuals in a line. For the rst position you have ten possible choices. Once the rst position is taken, you have 9 remaining options for the second place. So for the rst two positions you have 10 9 = 90 possibilities. For the third position there are only 8 possible guys to pick and so on. So for a line of 10 guys we have 10 9 8...2 1 = 10! options. A very used property of permutations is that n! = n (n 1)!. This is trivial to show by construction from the denition and is therefore left as an exercise. The intuition is the same as before. To compute the possible rearrangements of n elements we have n options for the rst place, and for each of those n choices we have all the permutations of the remaining (n 1) elements. The permutation of elements is the basis of binomial coecients, also called

combinations .
k
,

D enition 86 The combination of n elements, taken in groups of k is denoted by n n n

and is computed as
k

nk )! k! .

The rationale for this denition is the following. Arrange the n elements in a line. We will say that we obtain the same combination of k elements whenever the same k elements occupy the rst k positions in the line. This is very important, when we consider combinations, the ordering of the elements is immaterial. We already know that the total number of permutations of the n elements is n! . However, from all those n! arrangements, many of them correspond to the same combination, meaning that the same k objects occupy the rst k positions. If each combination is counted N times by computing n!, we then know that the true number of combinations ! will be n N. The remaining point is to nd N . To do this, just take a given arrangement and compute all the permutations that give the same combination of k elements. We have that for a given position of the last (n k ) elements, we can rearrange the rst

Appendix: Binomial Coecients

33

n This leads to the nal result that k

k elements in k ! ways; and in a similar fashion, for a given position of the rst k elements, we can rearrange the last (n k ) elements in (n k )! ways. Therefore the total number of rearrangements that give the same combination is N = (n k )! k !.

Exercise 87 Computing the factorial of a number may be cumbersome if n is large

n k
if

n! nk)! k! .

( try 1000! for example), but show that when computing combinatorial numbers this diculty may be overcome by calculating
+ 2) (n k + 1) = n (n 1) ... (n kk !
1000 4
.

is not too large. Just to convince yourself, compute

Example 88
52! 48! 4! .

Our rst use of binomial coecients will be to provide a guess for the binomial formula, that we have proven by induction above. In general, this is a limitation of the proofs by induction. We have to come up with the result before the proof begins. To nd a guess of the binomial formula we know that (a + b)n = (a + b) (a + b) (a + b) ... (a + b)
n times by the distributive law of multiplication, we know that each term of the binomial sum is calculated as the multiplication of n elements, some of them being a and being the rest b. It is therefore obvious that the binomial sum will have the form k n k (a + b)n = n . The only problem is to nd the k . But that is easy if k=0 k a b

The former is simply 1, since there is only one combination that comprises a 1 1 6 poker of aces. Therefore the chance is (52 ) = 270,725 3.7 10
4

Suppose you pick four cards from the top of a 52 card pile . What is the chance of your getting a poker of aces? The probability is computed as the number of favorable events divided by the total number of events. The latter is simply 52 = 4

we use combinatorial numbers. We can think of the term ak bnk as a list of n ordered elements, with k as and (n k) bs. For example if k = 1, then we know that there is only one a, either
in position 1,2, 3,..., n and bs. This is the same as computing in how many dierent positions can we nd

all we have to do is to nd in how many dierent ways can we rearrange the and (n

1 , n and the rest are b s. In general, for a xed particular

as k k
= k

as

taken in groups of

k) b s, which is then number of combinations of the n available positions

k.
So

, and the binomial formula is computed as n k k=0

n

(a + b)n =

ak bnk

34

Elementary Set Theory

2.6 References S.R. Lay, Analysis with an Introduction to Proof . Chapter 2. Third Edition. Prentice Hall. A. Matozzi, Lecture Notes Econ 897 University of Pennsylvania Summer 2001. A. N. Mitchell and C. J. Herget, Applied Algebra and Functional Analysis. Dover Publications. H.L. Royden, Real Analysis . Third Edition. Prentice Hall.