1 Set Theory and Functions

1.1 Basic Denitions and Notation

A set A is a collection of objects of any kind.
We write a A to indicate that a is an element of A: We express this as a is contained
in A.
We write A B if every element of A also is an element in B . We express this as A
is a subset of B
We write A = B if A and B consist of exactly the same elements. Otherwise we write
A ,= B:
If A B and A ,= B we say that A is a proper subset of B:
We write ? for the empty set, the set that contains no elements at all.
1.2 Operations on Sets
1. The union: Let A and B be two sets. Then, the union, denoted A ' B is the set of
all elements belonging to at least one of the two sets. That is,
A ' B = x[x A or x B :
Given an arbitrary collection of sets A
i
; where i is some parameter such that i I
(allows for nite or innite numbers) we write '
i
A
i
for the union, which is the set of
all elements that belongs to at least one of the sets in the collection,
'
i
A
i
= x[ x A
i
for some i I :
2. The Intersection: Given A and B we write A B for its intersection- the set of all
elements belonging to both A and B:
A B = x[x A and x B :
1
The generalization for a collection of sets A
i
is obviously

i
A
i
= x[ x A
i
for every i I
3. Disjoint Sets: A and B are said to be disjoint if they have no elements in common.
That is, if
A B = ?:
4. The Complement (sometimes called dierence): We (KF uses dierent notation)
write AB for the set of all elements in A that dont belong to B;
AB = x[x A and x = B
Remark 1 It follows immediately from the denitions that ' and satisfy communicative
and associative laws,
A ' B = B ' A
A B = B A
(A ' B) ' C = A ' (B ' C)
(A B) C = A (B C)
Hence, we can simply write A ' B ' C:
In addition:
Proposition 1 ' and satisfy the distributive law
(A ' B) C = (A C) ' (B C)
(A B) ' C = (A ' C) (B ' C)
Proof. Exercise on Problem Set 1
Sometimes we will also use the Cartesian product:
Denition 1 Given non empty sets A and B; the Cartesian product, denoted AB; is the
set of all ordered pairs (a; b) such that a A and b B
2
Example 1 If A = a
1
; a
2
; a
3
; a
4
and B = b
1
; b
2
then
A B = (a
1
; b
1
) ; (a
1
; b
2
) ; (a
2
; b
1
) ; (a
2
; b
2
) ; (a
3
; b
1
) ; (a
3
; b
2
) ; (a
4
; b
1
) ; (a
4
; b
2
)
Example 2 If A = [0; 1] and B = [0; 1] we write A B = [0; 1] [0; 1] or A B = [0; 1]
2
;
where geometrically the Cartesian product of the two (unit) intervals is the (unit) square.
1.3 Functions
The terminology Ill use is as follows:
Denition 2 A function (or mapping) from a set X to Y (denoted f : X Y ) is a rule
that assigns a unique element y = f (x) Y to every x X:
Usually, X is referred to as the domain of f:
Remark 2 Sometimes functions are dened in terms of the graph G, which is a collection
of ordered pairs x; f (x) X Y:We say that G is the graph of a function if and only if
for every x X there is a unique pair x; f (x) G.
We call the element f (x) Y the image of x (under f). This is generalized to subsets
as follows:
Denition 3 If A X we say that
f (A) = y Y [y f (a) for some a A
is the (direct) image of A:
Also,
Denition 4 If B Y we say that
f
1
(B) = x X[f (x) B
is the inverse image (or preimage) of B:
3
The image and the inverse image can be used to dene some important types of functions:
Denition 5 f : X Y is said to be surjective (onto) if f (X) = Y
Denition 6 f : X Y is said to be injective (one to one) if f (x
1
) ,= f (x
2
) for every pair
(x
1
; x
2
) X such that x
1
,= x
2
:
Denition 7 f : X Y is said to be bijective if it is surjective and injective.
It is more or less immediate that:
Proposition 2 If f is a bijection, then there is a (uniquely dened) function g : Y X
dened by letting g (y) X be given by g (y) = f
1
(y) for every y Y: This function is
called the inverse of f (since no confusion can arise we will use f
1
also to denote the inverse
function in the rest of the course).
In some sense the conclusion is trivial. However, well write down a proof to make sure
we understand what the denitions mean
Proof. Since f is injective there are two possibilities:
1. f
1
(y) = ?
2. there exists a unique element x X such that x = f
1
(y)
To see this, assume that for contradiction that there exists two distinct elements x
1
and
x
2
such that x
1
f
1
(y) and x
2
f
1
(y) : Then,
x
1
; x
2
f
1
(y) = x X[f (x) y = x X[f (x) = y
which contradicts the assumption that f is injective. Moreover, since f is surjective
Y = f (X) = y Y [y f (x) for some x X ;
which can be rephrased as saying that there exists some x X such that f (x) = y for every
y Y: This rules out f
1
(y) = ?; so the result follows
4
Denition 8 If f : X Y and g : Y Z the composition f g is given by the function
h : X Y such that h(x) = g (f (x)) for every x X:
Proposition 3 Suppose that f : X Y and g : Y Z are bijective, then the composition
f g is bijective.
Proof. Fix any z
0
Z: Since g : Y Z is surjective there exists y
0
Y such that g (y
0
) = z
0
:
Since f is surjective there exists x
0
X such that f (x
0
) = y
0
: Hence,
g (f (x
0
)) = g (y
0
) = z
0
Since z
0
Z was arbitrary g (f (X)) = Z:
Suppose that f g is not injective. Then there exists x
1
,= x
2
such that g (f (x
1
)) =
g (f (x
2
)) : Since f is injective f (x
1
) = y
1
,= y
2
= f (x
2
) : Hence, there exists y
1
,= y
2
such
that g (y
1
) = g (y
2
) contradicting that g is an injection.
1.4 Sequences I
Just like functions can be dened in terms of ordered pairs (dening the function from its
graph) we may dene an ordered pair as a function
f : 1; 2 X:
Similarly, a nite sequence can be dened as a mapping
f : 1; ::::; n X:
Now, letting N be the set of natural numbers we can dene: an innite sequence as a mapping
f : N X:
It is customary to depart somewhat from standard functional notation when dealing with
sequences and,
1. Write x
i

n
i=1
instead of (f (1) ; f (2) ; :::; f (n)) or f (i)
n
i=1
2. Write x
i

1
i=1
instead of f (i)
1
i=1
5
1.5 Finite, Innite, Countable and Uncountable Sets
1.5.1 Finite vs Innite Sets
A nite set is a set which is either empty or a set for which there exists some natural number
n such that the set has exactly n elements. That is:
Denition 9 The set A is nite if it is empty or if there exists some n N and a bijection
f : 1; :::; n A (this denes a set with exactly n elements too): A set A is innite if it is
not nite.
Remark 3 Since the bijection f : 1; :::; n A denes an inverse f
1
: A 1; :::; n
niteness could be dened as a bijection from A onto 1; :::; n : Both conventions are used
to dene niteness (and similarly for countability).
The remark above also proves part of the following useful claim:
Proposition 4 A set A is nite if and only if there exists a nite set B and a bijection
g : A B
Proof. (==) If A is nite there exists some nite n and a bijection f : 1; :::; n A;
implying that g = f
1
: A 1; :::; n is well-dened. Letting B = 1; :::; n proves the rst
part.
(==) Suppose that there exists nite set B and a bijection g : A B: Since B is nite
there exists some n N and (using the remark) a bijection h : B 1; :::; n : Since g and h
are bijective g h : A 1; ::::; n is bijective. Hence, there is a bijection f : 1; :::; n A
where
f (i) = g h(i) for i = 1; ::::; n
We now prove a result sometimes used in combinatorics, which is referred to as the
Pigeonhole principle. If there are more pigeons than pigeonholes at least two pigeons need
to share a hole.
6
Proposition 5 Suppose that m; n N: Then:
1. If m _ n there exists an injection (onto map) f : 1; :::; m 1; ::; n
2. If m > n there is no injection f : 1; :::; m 1; ::; n :
Proof. For part 1, let f : 1; :::; m 1; ::; n be given by
f (j) = j for every j 1; :::; m _ 1; ::; n :
Obviously, f is injective
For the part 2, we begin by noting that if n = 1; then the only map from f : 1; :::; m
1; ::; n is f (j) = 1 for every j 1; :::; m ; which is not an injection for any m > 1.
Next, assume that there is no injection f : 1; :::; m 1; ::; k for any m > k: We need
to demonstrate that it follows that there is no injection f : 1; :::; m 1; ::; k + 1 for any
m > k + 1; which will prove the claim by induction.
For contradiction, suppose such an injection would exist. Then, there must be some
j 1; :::; m such that f (j) = k +1 since otherwise h : 1; :::; m 1; ::; k is an injection
despite m > k + 1 > k; which violates the induction hypothesis. Moreover, it must be the
case that exactly one element j

1; :::; m is such that f (j) = k +1 since obviously there

is no injection more than one element is mapped into k + 1 : It follows that the function

h : 1; :::; m j

1; ::; k
dened as

h(j) = h(j) for every j 1; :::; m j

must be an injection. At this point we may either just assert that 1; :::; m j

and
1; :::; m1 are equivalent sets or relabel by constructing the map g : 1; :::; m1
1; ::; k
g (j) =
_
_
_

h(j) = h(j) if j < j

h(j 1) = h(j 1) if j > j

:
7
It is obvious that g is injective if and only if

h is injective, but g is not injective according
to the induction hypothesis. The result follows
We can extend this result to
Proposition 6 Suppose that m N: Then:
1. There exists an injection (onto map) f : 1; :::; m N
2. There is no injection f : N 1; ::; m
Proof. For part 1, let f : 1; :::; m N be given by
f (j) = j for every j 1; :::; m _ N:
Obviously, f is injective
For part 2, if f : N 1; ::; m is an injection it follows that f (j) ,= f (i) for every
pair i; j N; which implies that f (j) ,= f (i) for every pair i; j 1; :::; m + 1 ; which
contradicts the previous proposition
I will take the following for granted (see Bartle and Sherbert pages 61-62)
1.5.2 Countable vs. Uncountable Sets
Countably innite sets can be dened in similar fashion to the denition of nite sets in
Denition 9. However, it is more convenient to use the following denition:
Denition 10 A set A is said to be countable if there exists an injective function f : A
N:
Bartle and Sherbert use a dierent denition and establish existence of an injection
f : A N as a proposition.
Example 3 Let A = N: Consider the identity function f : N N given by
f (i) = i for every i N:
Obviously, this is injective, so N is countable.
8
Example 4 Let Z = ::: 1; 0; 1; ::: be the set of positive and negative integers. Let f :
Z N be given by
f (i) =
_
_
_
2i if i < 0
2i + 1 if i _ 0
;
which clearly is injective (as negatives map into even and positives map into odd numbers).
Example 5 Let A be nite. Then there is some n N and a bijection f : A 1; :::; n :
Dening f
1
: A N as
f
1
(a) = f (a)
for each a A is obvious that f
1
is injective (but not surjective as there exists no a A such
that f
1
(a) = j when j > n). Hence, any nite set is countable
Combining with previous results we conclude that:
Proposition 7 Let A be a countable set. Then, A is (countably) innite if and only if there
exists some bijective f : A N:
Proof. (=) For contradiction, suppose f is surjective and that A is nite. If f is bijective
f
1
: N A exists and is injective (and surjective). If A is nite there exists n N and
bijection g : A 1; :::; n : Hence f
1
g : N 1; :::; n is injective (and surjective),
which contradicts Proposition 6. Hence, A cannot be nite, implying that it is innite
(=)Suppose that A is nite and that there exists a bijection f : A N: Then
f
1
: N A
is bijective. Moreover, since A is nite there exists a bijection g : A 1; :::m ; where
m N: Hence,
f
1
g : N 1; :::; n
is bijective, which is impossible due to Proposition 6.
Theorem 1 If A is countable and B A; then B is countable (if A is nite, then B is
nite).
9
Proof. First, consider the case with A nite. Without loss, suppose B is non-empty (empty
set if nite and therefore countable by default). If B is innite there exists bijection
f : B N
Let b
0
B A and consider map g : A : B
g (a) =
_
_
_
a if a B
b
0
if a = B
:
As this map restricted on B is the identity map, this is obviously surjective. Moreover, since
A is nite there exists n and bijective map h : 1; :::; n A. Hence
h g f : 1; :::; n N
is surjective, which which is impossible due to Proposition 6.
Now, suppose A is countably innite so that there exists bijection f : A N: Let
g : B A be given by
g (b) = b
for every b B: This is obviously injective, so g f : B N is injective.
Theorem 2 The union of a countable number of countable sets is countable.
Proof. Proof is very similar to proof of countability of rationals.
Without loss, suppose that A
i

i
is a collection of disjoint sets (otherwise look at
A
1
; A
1
A
2
; A
3
(A
1
' A
2
) :::::).
. Write a
ij

j
for the elements. Let f : '
i
A
i
N be given by
f (a
ij
) = 2
i
3
f
i
(a
ij
)
:
Which is well dened and injective.
[If not there is some a
ij
,= a
kl
with
2
i
3
f
i
(a
ij
)
= 2
k
3
f
k
(a
kl
)
=
2
ik
= 3
f
k
(a
kl
)f
i
(a
ij
)
10
LHS values
::::
1
16
;
1
8
;
1
4
;
1
2
; 1; 2; 4; 8; 16::::
RHS values
1
81
;
1
27
;
1
9
;
1
3
; 1; 3; 9; 27; 81; :::
Theorem 3 The set of rationals is countable.
Proof is very similar.
Theorem 4 The set of real numbers in the closed interval [0; 1] is uncountable.
Proof. If [0; 1] is countable we could nd an injection f : N [0; 1] : Let x
i

i2N
denote
the implied sequence of elements in [0; 1] and write them in decimal form as
x
1
= f (1) = 0:x
11
x
12
::::::x
1n
::::::::
x
2
= f (2) = 0:x
21
x
22
::::::x
2n
::::::::
::::
x
n
= f (n) = 0:x
n1
x
n2
::::::x
nn
::::::::
:::::
Consider
y = 0:y
1
y
2
::::y
n
::::::
where every y
i
= 0; x
ii
; 9 [ruling out 0 and 9 is to avoid dealing with 0:5 versus 0:499999::::]
Hence, y does not belong to x
i

i2N
: Since the sequence was arbitrary, no sequence contains
all the reals, which completes the proof.
11
2 Sequences and Limits in R
2.1 Suprema and Inma
We will not be careful about the axioms of the real number system, but rather proceed based
on our intuitive understanding of ordering properties and arithmetic. However, we will be
explicit about ONE crucial property, which relates to a more general completeness property
for metric spaces.
Denition 11 Given a nonempty set A R we say that u is an upper bound for A if x _ u
for every x A:
Denition 12 If an upper bound for A exists we say that A is bounded above.
Lower bounds/bounded below are dened the same way.
Denition 13 Given a nonempty set A R we say that u is a least upper bound for A if:
1. u is an upper bound for A
2. x _ u for any upper bound x for A
Example 6 The set of reals, R; has no upper or lower bound.
Example 7 Any x _ 0 is an upper bound for (; 0[ : The unique least upper bound for
(; 0[ is 0:
Example 8 Any x _ 0 is an upper bound for (; 0) : The unique least upper bound for
(; 0[ is 0:
Proposition 8 If a least upper bound to A exists, then it is unique.
Proof. Suppose that u and u
0
are both least upper bounds. Then, u _ u
0
since otherwise u
wouldnt be a least upper bound and u
0
_ u for the same reason. Hence u = u
0
In what follows we will call the least upper bound to a subset of R the supremum of
the set. The most crucial assumption about the reals is:
12
Axiom 1 (Supremum property of R) For any non empty subset A of R which is bounded
above the supremum exists.
Idea is that there are no gaps among the set of reals.
To appreciate that the axiom is a bit more subtle than one way rst think consider the
example:
Example 9 Let Q be the set of rational numbers and A = x Q[x
2
< 2 : Intuitively we
then notice that
1. if u is an upper bound for A; then u
2
= 2: (proof to follow below)
2. there is no rational number such that u
2
= 2 (proof to follow below)
A consequence of the supremum property of the reals is that the innum exists for sets
that are bounded below.
Proposition 9 If a nonempty subset of the reals A is bounded below, then it has a greatest
lower bound (innum).
Proof. Problem set. Idea is that l is a lower bound of A if and only if l is an upper bound
for
x R[ x A
We can then prove Archimedes axiom which can be thought of as either of the following
two statements:
(1) Given any number, you can always pick an integer that is larger than the original
number.
(2) Given any positive number, you can always pick an integer whose reciprocal is less
than the original number.
Proposition 10 The set of natural numbers N is not bounded above.
13
Proof. If N is bounded above, then u = sup N exists by the completeness axiom. Let
n N be picked arbitrary. We have that n _ u as N is bounded above. But, n + 1 is also
bounded by u; so
n + 1 _ u = n _ u 1:
Since the argument holds for an arbitrary n N it follows that u1 is also an upper bound
for N; which contradicts the leastness of u:
Corollary 1 For any pair (x; y) R such that x ,= y there exists a rational number in
between.
Proof. Without loss, assume that x < y: Suppose that x _ 0: By Proposition 10 there
exists n such that
n >
1
y x
;
implying that
1
n
< y x
Moreover, there exists v N such that (here we use fact that x _ 0)
v 1 _ nx < v
since:
1. Set m N[m _ nx is nonempty (otherwise nx upper bound for N)
2. Set m N[m _ nx is bounded below by nx =exists a least element v m N[m _ nx
x <
v
n
_ x +
1
n
< x +y x = y:
Now, let x < 0: Suppose that y > 0: Then, the previous argument establishes that
there exists n; v N such that
x < 0 <
v
n
< y:
14
Finally, let x < y < 0: Then, the previous argument establishes that there exists n; v
such that
y <
v
n
< x;
implying that
x <
v
n
< y:
We can now get back to the example:
Example 10 (continued) CLAIM: if u is a least upper bound for A = x Q[x
2
< 2
then u
2
= 2:
Proof. Obviously an upper bound exists (20 works...and is in Q).
CASE 1: Suppose the least upper bound is such that u
2
< 2: By the Archimedan property
just proved we may pick n such that
0 <
1
n
<
2 u
2
4u
=
4u
n
< 2 u
2
and
1
n
< 2u;
But then (observing that we can trivially rule out the possibility that u is negative)
u
2
<
_
u +
1
n
_
2
= u
2
+
2u
n
+
1
n
2
_
1
n
< 2u
_
< u
2
+
2u
n
+
2u
n
= u
2
+
4u
n
_
4u
n
< 2 u
2
_
< u
2
+ 2 u
2
= 2;
implying that u is not a lower bound.
CASE 2: Suppose that u
2
> 2 . By the Archimedan property we can pick n such that
0 <
1
n
<
u
2
2
2u
=
_
u
1
n
_
2
= u
2

2u
n
+
1
n
2
> u
2

2u
n
> u
2

_
u
2
2
_
= 2;
implying that u
1
n
is a smaller upper bound of A; contradicting the leastness of u:
15
Remark 4 CHECK: SUM OF RATIONAL S IS RATIONAL. HENCE THE EXAMPLE
SHOWS (ACCEPTING THAT
_
2 IS IRRATIONAL) THAT THE SUPREMUM PROP-
ERTY DOES NOT APPLY TO THE SET OF RATIONAL NUMBERS.
Remark 5 By the supremum property of R the example also shows that
_
2 exists in the
real number system.
2.2 Real Sequences
While formally a real sequence is a map f : N R; but we will usually write x
1
; x
2
; :::; x
i
; ::::
or x
n

1
n=1
when this cannot cause any confusion.
A crucial concept is that of convergence:
Denition 14 x
i

1
i=1
converges to the (real number) x

if given any (real number) " > 0

there exists some nite K (possibly dependent on ") such that
x

" < x
n
< x

+"
for every n _ K:
Example 11 Let x
n

1
n=1
be given by x
n
=
n1
n
for every n N: By the Archimedean
property of the naturals there exists K such that
0 <
1
K
< "
for every " > 0: Hence, if n _ K
[x
n
1[ =

n 1
n
1

1
n

=
1
n
_
1
K
:
We conclude that x
n
=
n1
n
converges to 1:
Example 12 Sometimes limit not so easy to guess, as with
x
n
=
_
1
1
n
_
n
16
This can be motivated from probability theory: The chance of of k successes in n Bernoulli
trials is
p
kjn
=
n!
k! (n k)!
p
k
(1 p)
nk
So, if the probability of success is
1
n
; then the probability of no success out of n draws is
n!
1 n!!
_
1
n
_
0
_
1
1
n
_
n
=
_
1
1
n
_
n
:
Can show that
_
1
1
n
_
n

1
e
:
Proposition 11 If x
n

1
n=1
converges there is a unique limit.
Proof. Suppose that x

and x

are limits of x
n

1
n=1
and let x

< x

without loss of
generality. Let " =
x

3
: Because x

is a limit of x
n

1
n=1
there exists K

such that
x
n
< x

+" = x

+
x

3
=
2
3
x

+
1
3
x

for all n _ K

: Symmetrically, as x

is a limit of x
n

1
n=1
there exists K

such that
x
n
> x

" = x

3
=
2
3
x

+
1
3
x

Hence,
2
3
x

+
1
3
x

< x
n
<
2
3
x

+
1
3
x

=
2
3
(x

) <
1
3
(x

) :
= 2 < 1:
Denition 15 A sequence of real numbers x
n

1
n=1
is:
1. monotonic increasing if x
n+1
_ x
n
for every n N
2. monotonic decreasing if x
n+1
_ x
n
for every n N
3. monotonic if either monotonic increasing or decreasing
17
Denition 16 A sequence of real numbers x
n

1
n=1
is bounded if there exists a; b R such
that a _ x
n
_ b for every n N:
Proposition 12 Every bounded monotonic sequence is convergent.
Proof. Let X = x
1
; x
2
:::; x
n
; :::: R be the set of elements of the sequence x
n

1
n=1
Remark 6 Formally, a real valued sequence is dened as a function f : N R; so X =
f (N) ; the direct image of N
Consider the case where the sequence is increasing. Then, X is bounded above by a: By
the completeness/supremum property of the real number system there exists a supremum of
X: Write x = sup X:
Consider an arbitrary " > 0 and note that x " can not be an upper bound of X; since
that would contradict that x is the LEAST upper bound. Hence, there exists K < such
that x
K
> x "; which implies that
x _ x
n
_ x
K
> x "
for every n _ K: Hence
[x
n
x[ _ "
for every n _ K: Since " was arbitrary, x is the limit of x
n

1
n=1
:
Decreasing sequences-same argument.
2.3 Subsequences
Denition 17 Let x
n

1
n=1
be a sequence and k
i

1
i=1
a strictly monotonic sequence with
k
i
N for every i: Then, the sequence
x
kn

kn
= x
k
1
; x
k
2
; :::::; x
kn

(constructed by setting x
k
i
= x
n
for every k
i
) is called a subsequence of x
n

1
n=1
18
Example 13 Let x
n

1
n=1
be given by
x
n
=
_
_
_
a if n is odd
b if n is even
;
then two obvious subsequences are (a; a; :::; a; :::) and (b; b; :::; b; )
Theorem 5 (Bolzano & Weierstrass) Every bounded sequence of real numbers has a
convergent subsequence.
Proof. Let a
0
and b
0
be bounds for x
n

1
n=1
, so that
x
n
[a
0
; b
0
]
for every n N: Now let
L
1
=
_
a
0
;
a
0
+b
0
2
_
and let N
L
1
=
_
n N[x
n

_
a
0
;
a
0
+b
0
2
__
H
1
=
_
a
0
+b
0
2
; b
0
_
and let N
H
1
=
_
n N[x
n

_
a
0
+b
0
2
; b
0
__
There are now two possibilities:
1. N
L
1
is innite. If this is the case we let a
1
= a
0
and b
1
=
a
0
+b
0
2
and let N
1
=
_
n N[x
n

_
a
0
;
a
0
+b
0
2
_
2. N
L
1
is nite. If this is the case it follows that N
H
1
is innite. If this is the case we let
a
1
=
a
0
+b
0
2
and b
1
= b
0
, and let N
1
= N
H
1
:
We conclude that there exists an interval [a
1
; b
1
] [a
0
; b
0
] and an innite set N
1
=
n N[x
n
[a
1
; b
1
], where
b
1
a
1
=
1
2
[b
0
a
0
]
We can then construct [a
2
; b
2
] ; [a
3
; b
3
] :::: and N
2
; N
3
::: by continuing to divide the intervals
in two halfs. We then have::
CLAIM: For every k N there exists an interval [a
k
; b
k
] with
b
k
a
k
=
1
2
k
[b
0
a
0
]
19
and an innite set
N
1
= n N[x
n
[a
k
; b
k
]
:Suppose claim is true for k N: Let
L
k+1
=
_
a
k
;
a
k
+b
k
2
_
and let N
L
k+1
=
_
n N[x
n

_
a
k
;
a
k
+b
k
2
__
H
k+1
=
_
a
k
+b
k
2
; b
k
_
and let N
H
k+1
=
_
n N[x
n

_
a
k
+b
k
2
; b
k
__
Since
N
L
k+1
' N
H
k+1
=
_
n N[x
n

_
a
k
;
a
k
+b
k
2
__
'
_
n N[x
n

_
a
k
+b
0
2
; b
k
__
= n N[x
n
[a
k
; b
k
]
is innite by the induction hypothesis we conclude that at either N
L
k+1
or N
H
k+1
is innite.
As
a
k
+b
k
2
a
k
= b
k

a
k
+b
k
2
=
1
2
(b
k
a
k
) =
1
2
k
[b
0
a
0
]
by the induction hypothesis this proves the claim. Now:
1. a
k
[a
0
; b
0
] for every k and a
k
_ a
k1
for every k =limit a

exists (monotone bounded

convergence result)
2. b
k
[a
0
; b
0
] for every k and b
k
_ b
k1
for every k =limit b

exists (monotone bounded

convergence result)
3. c
k
= b
k
a
k
[0; b
0
a
0
] for every k and c
k+1
= b
k+1
a
k+1
=
1
2
(b
k
a
k
) _ (b
k
a
k
) =
c
k
=limit c

exists (again)
4. c

= 0:
5. c

= b

: Hence both a
k
and b
k
has the same limit a

= b

:
20
Now, pick " > 0, let (very inecient choice)
k inf
_
n N[n _
2 (b
0
a
0
)
"
_
=
1
k
_
"
2 (b
0
a
0
)
And let
K (") = inf n N[x
n
[a
k
; b
k
]
Then for every n _ K (") it follows that
x
n
a

_ b
k
a

_ b
k
a
k
=
1
2
k
[b
0
a
0
]
<
1
k
[b
0
a
0
] _
"
2 (b
0
a
0
)
[b
0
a
0
] =
"
2
;
which proves the result.
2.4 Open and Closed Sets in R
Denition 18 Given any " > 0 we say that an open ball (" neighborhood) of x R is given
by the set
B(x; ") = y R[ [x y[ < " = (x "; x +") :
Denition 19 A set A R is open if for every x A there exists an open ball B(x; ") A:
Denition 20 A set A R is closed if the complement B = RA is open.
Example 14 Let A = R: Fix x and let " = 1: Then any y (x 1; x + 1) R; so R is
open.
Example 15 Consider the open interval A = (a; b) :Pick any x (a; b) and let
" = min x a; b x
21
then for any y B(x; ")
y > x " = x min x a; b x
_ x + max a x; x b _ x +a x = a
y < x +" = x + min x a; b x
_ x +b x = b:
Hence, (a; b) is an open set.
Example 16 Let A = (; a) and B = (b; ) : Open, by same argument.
Example 17 Consider [a; b] : Not open since for every " > 0
B(a; ") [a; b] = (a "; a]
in nonempty for every " > 0:
Note that
R [a; b] = (; a) ' (b; ) :
By the previous argument, if x (; a) we can nd a ball B(x; ") (; a) and if x is
in (b; ) we can also nd a ball B(x; ") (b; ) : Hence, there an open ball B(x; ")
(; a) ' (b; ) for every x (; a) ' (b; ) ; so (; a) ' (b; ) : we conclude that [a; b] is
a closed set.
Example 18 ? is open (vacuously since there is no point in the set). Hence, R is closed
and open. Since R is also open, it follows that ? is open and closed.
Theorem 6 The union of any collection of open sets is open.
Proof. Let I be some indexing set and assume that A
i
is open for every i I: If x A =
'
i2I
A
i
: Then there exists j I and " > 0 such that B(x; ") A
j
: Hence,
B(x; ") A
j
'
i2I
A
i
:
22
Theorem 7 The intersection of a nite collection of open sets is open.
Proof. Let x
k
i=1
A
i
: Let "
i
> 0 be such that B(x; "
i
) A
i
; which exists since x

k
i=1
A
i
= x A
i
for every i and each A
i
is open. Let
" = min "
1
; "
2
; ::; "
k
> 0;
and note that
B(x; ") B(x; "
i
) A
i
for every i
implying that
B(x; ")
k
i=1
A
i
:
Example 19 Let
_
0;
1
k
_
and consider

k
i=1
_

1
k
; 1 +
1
k
_
=
_

1
k
; 1 +
1
k
_
which is open for every nite k: However

1
i=1
_

1
k
; 1 +
1
k
_
= [0; 1] ;
which is not open.
Theorem 8 The intersection of an arbitrary collection of closed sets is closed.
Proof. Let I be some indexing set and assume that A
i
is closed for every i I: By de
Morgans laws
R (
i2I
A
i
) = '
i2I
RA
i
:
Each set RA
i
is open, so (by previous theorem) '
i2I
RA
i
is open, proving that R (
i2I
A
i
)
is open, which by denition means that
i2I
A
i
is closed.
Theorem 9 The union of a nite collection of closed sets is closed.
23
Proof. By de Morgans laws
R
_
'
k
i=1
A
i
_
=
k
i=1
RA
i
:
Each set RA
i
is open and k is nite, so (by previous theorem)
k
i=1
RA
i
is open, proving
that R
_
'
k
i=1
A
i
_
is open, which by denition means that '
k
i=1
A
i
is closed.
Denition 21 (ALT 1) Given set A R; x R is called a cluster point of A if for every
" > 0 there exists y B(x; ") A x.
Proposition 13 A set A is closed if and only if it contains all its cluster points.
Proof. (=) Suppose not. Then, there exists an cluster point x RA; which is an open
set. Hence, there exists " > 0 such that B(x; ") RA, implying that B(x; ") A = ?;
contradicting that x is a cluster point of A:
(=) Suppose that A contains all cluster points. Pick y RA: Then, y is not a cluster
point, so there exists open ball B(y; ") RA:True for all y RA; so RA is open. Hence,
A is closed.
Almost the same result stated in terms of sequences.
Proposition 14 A is closed if and only if the limit of every converging sequence x
n
with
x
n
A for all n belongs to A:
Proof. We will prove this from the previous result by showing that
CLAIM: A contains all its cluster points=the limit of every converging sequence x
n

with x
n
A for all n belongs to A:
(=) Suppose that x
n
converges to x

:Then, for every " > 0 there exists K such that

x
n
B(x

; ") for n _ K; and, by hypothesis, x

n
A: Hence, x

is a cluster point of A: By
assumption A contains its cluster points, so lim
n!1
x
n
= x

A:
Suppose that x

= A is a cluster point. Let "

1
> 0 and pick x
1
B(x

; "
1
) x

arbitrarily. As x
1
,= x

there exists "

2
> 0 such that x
1
= B(x

; "
1
) ; implying that there
exists x
2
B(x

; "
2
) B(x

; "
1
) because x

is a cluster point: Recursively, if x

k
,= x

24
and x
k
B(x

; "
k
) there exists "
k+1
> 0 such that x
k
= B(x

; "
1
) : Hence, there exists
x
k+1
B(x

; "
k+1
) B(x

; "
1
) : This denes an innite sequence which obviously converges
to x

; so x

A if the limit of every converging sequence is in A:

An alternative denition is.
Remark 7 The proof shows that given any set A R; x R is a cluster point of A if for
every " > 0 there exists an innite set X B(x; ") A. This is an alternative denition
often used for a cluster point.
2.5 Compact Sets
Denition 22 An open cover of a set A in R is a collection ( = C
i
of sets such that
A '
i2I
C
i
Denition 23 (
0
is a subcover of ( if each set C
j
in (
0
also is in ( and (
0
is a cover of A:
Denition 24 A in R is said to be compact if every open cover ( has a nite open subcov-
ering (
0
:
Theorem 10 (Heine Borel) A R is compact if and only if A is closed and bounded.
Remark 8 Every analysis book says that result is "closely related to Bolzano Weierstrass",
yet proofs tend to not use that result. Here is an argument that makes explicit use of Bolzano
Weierstrass.
Proof. (=) Suppose that A is compact. We note that if C
i
= (i; i) then
'
1
i=1
C
i
is an open cover of R: Hence,
A R '
1
1
C
i
so (i; i) [i Nis an open cover of A as well. Since it is compact, there exists some nite
K such that
A '
K
i=1
C
i
= (K; K) ;
25
so A is bounded. Let y = A and let
C
i
= R
_
y
1
i
; y +
1
i
_
Obviously, G
i
is open (
_
y
1
i
; y +
1
i

is closed) and we have that '

1
i=1
C
i
= R which covers
R and therefore also A: By compactness there is some nite K such that
A '
K
i=1
C
i
= R
_
y
1
K
; y +
1
K
_
=
A
_
y
1
K
; y +
1
K
_
= ? =
A
_
y
1
K
; y +
1
K
_
= ?
Hence, for every y = A = y RA there exists " > 0 such that
_
y
1
K
; y +
1
K
_
RA; so
RA is open. Hence, A is closed.
(=) For contradiction, suppose A is closed and bounded, but that that there is an open
cover ( = C

for which there is no nite subcover.

For any n, consider the collection of open sets
(
n
= C
;n
where C
;n
=
_
x C

[ inf
y2RnC
[x y[ >
1
n
_
CASE 1: Suppose that there exists K such that (
K
is a subcover of (. Since A is bounded
by there exists m such that A [m; m] so let
x
1
= m and B
_
x
1
;
1
K
_
=
_
m
1
K
; m+
1
K
_
x
2
= m+
1
2K
and B
_
x
2
;
1
K
_
=
_
m
1
2K
; m+
3
2K
_
::::
x
n
= m+
(n 1)
2K
and B
_
x
n
;
1
K
_
=
_
m+
(n 3)
2K
; m+
(n + 1)
2K
_
::::
x
4mK+1
= m+
(4mK + 1 1)
2K
= m+ 2m = m
Hence,
_
B
_
x
n
;
1
K
__
4mK+1
n=1
is a nite cover of A: Moreover, each x
n
C
n;K
by the hypothesis
that (
K
is a subcover of (. Pick any y B
_
x
n
;
1
K
_
: Then,
inf
x2Rnxn2C
n;K
[x y[ _ inf
xn2C
n;K
[x x
n
[ [y x
n
[ > 0
26
since [y x
n
[ <
1
K
as y is in B
_
x
n
;
1
K
_
and inf
xn2C
n;K
[x x
n
[ _
1
K
by construction of
C
;K
:
We conclude that B
_
x
n
;
1
K
_
C
n;K
: As
_
B
_
x
n
;
1
K
__
4mK+1
n=1
is an open cover it follows
that C
n;K

4mK+1
n=1
is an open cover of A: Obviously, C
n;K

4mK+1
n=1
is a nite subcover of
(
n
= C
;n
; and by assumption (
n
= C
;n
is a subcover of (: Hence, A is compact.
CASE 2: Suppose A is not covered by (
n
= C
;n
for any n: Then there is a
sequence x
n
with x
n
A and x
n
R '

C
n
for every n N: Let x
kn
be a convergent
subsequence with limit x

: Then we note that:

1. Since A is closed x

A
2. Since ( is a covering of A there exists

such that x

3. C

is open so there exists " > 0 such that B(x

; ") C

4. Since there is no n such that x

'

C
;n
there is no n such that
x

n
=
_
x C

[ inf
y2RnC

[x y[ >
1
n
_
= there is no n such that B
_
x

;
1
n
_
C

[if B
_
x

;
1
n
_
C

then [x

y[ _
1
n
>
1
n+1
for every y = C

implying that x

n+1
]
5. Statements there exists " > 0 such that B(x

; ") C

and there exists no n such

that B
_
x

;
1
n
_
C

contradict each other.

We conclude that there cannot be any convergent subsequence to x
n
; which (since
x
n
is bounded) contradicts the Bolzano Weierstrass theorem. It follows that A must be
covered by (
n
= C
;n
for some n; which by analysis in CASE 1 means that A is compact.
2.6 Cauchy Convergence
An alternative way of thinking about convergent sequences is as a sequence with terms that
get closer and closer together further out in the sequence.
27
Denition 25 x
n

1
n=1
is said to be a Cauchy sequence if given any " > 0 there exists K
such that
[x
n
x
m
[ < "
for every m; n _ K:
The set of real numbers is a simple example of a complete metric space. Such spaces
have many important properties, one being that the two convergence criteria are equivalent
(which in the end gives a nice characterization of compactness).
The easy direction is:
Lemma 1 Every convergent sequence of real numbers x
n

1
n=1
is a Cauchy sequence.
Proof. Fix " > 0 and let x
n

1
n=1
have limit x

: As x
n

1
n=1
converges to x

there exists
K
_
"
3
_
N such that
[x
n
x

[ <
"
2
[x
m
x

[ <
"
2
for every n; m _ K
_
"
2
_
: Thus (triangle inequality "direct path shorter than any other path")
[x
n
x
m
[ = [(x
n
x

) + (x
m
x

)[
_ [(x
n
x

)[ +[(x
m
x

)[
<
"
2
+
"
2
= "
To establish the other direction is a bit more work, but we may rely on some of the work
we have already done. In addition, the following fact is important.
Lemma 2 If x
n

1
n=1
is a Cauchy sequence, then there exists a; b such that a < x
n
< b for
every n:
28
Proof. Pick some " > 0 and let K be such that [x
n
x
m
[ < " for every n; m _ K: Then
[x
n
x
K
[ < "
for every n; implying that
x
n
x
K
< " and x
K
x
n
< "
x
K
" < x
n
< x
K
+"
for every n _ K: The set
x
1
; x
2
; :::; x
K
"; x
K
+"
is obviously bounded above and below and any bounds a; b to this set also bounds x
K
+":
Theorem 11 x
n

1
n=1
converges if and only if it is a Cauchy sequence.
Proof. Let x
n

1
n=1
be a Cauchy sequence. By Lemma above it is then bounded above and
below. Applying the Bolzano-Weierstrass Theorem we know that there exists a convergent
subsequence x
kn

kn
. Let the limit of the convergent subsequence be denoted by x

: Fix
" > 0: Since x
kn
converges to x

there exists K
1
k
i

1
i=1
such that
[x
n
x

[ <
"
2
for every n _ K
1
such that there exists an element in k
i

1
i=1
with k
i
= n: This implies that
[x
K
1
x

[ <
"
2
:
But, since x
n

1
n=1
is Cauchy there exists K
2
such that
[x
n
x
K
1
[ <
"
2
holds for every n _ K
2
: Therefore
[x
n
x

[ = [(x
n
x
K
1
) + (x
K
1
x

)[
_ [x
n
x
K
1
[ +[x
K
1
x

[ < "
holds for every n _ K
2
: We conclude that x

is the limit of x
n

1
n=1
:
29