Intro Avram

Lecture Notes
Basic Concepts of Analysis

MATH 372
Instructor: Ivan Avramidi
Textbook: R. G. Bartle and D. R. Sherbert, (John Wiley, 2000)
New Mexico Institute of Mining and Technology
Socorro, NM 87801
June 3, 2004
Author: Ivan Avramidi; File: introanal.tex; Date: November 21, 2005; Time: 18:02
Contents
1
Preliminaries
1.1 Sets and Functions . . . . . . . .
1.1.1 Set Operations . . . . . .
1.1.2 Cartesian Products . . . .
1.1.3 Mappings and Functions .
1.1.4 Direct and Inverse Images
1.1.5 Special Types of Functions
1.1.6 Inverse Functions . . . .
1.1.7 Composition of Functions
1.1.8 Restrictions of Functions
1.2 Finite and Infinite Sets . . . . . .
1.2.1 Countable Sets . . . . . .
1.2.2 Homework . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
The Real Numbers

2.1 The Algebraic and Order Properties of R
2.1.1 Rational and Irrational Numbers
2.1.2 The order Properties of R . . .
2.1.3 Inequalities . . . . . . . . . . .
2.2 Absolute Value and Real Line . . . . .
2.2.1 The Real Line . . . . . . . . .
2.2.2 Homework . . . . . . . . . . .
2.3 The Completeness Property of R . . . .
2.3.1 Suprema and Infima . . . . . .
2.3.2 The Completeness Property of R
2.3.3 Homework . . . . . . . . . . .
2.4 Applications of Supremum Property . .
2.4.1 Functions . . . . . . . . . . . .
I
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
. .
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
2
3
3
4
4
4
5
5
6
6
7
.
.
.
.
.
.
.
.
.
.
.
.
.
9
9
11
12
14
15
16
16
17
17
19
19
20
20
CONTENTS
II
2.5
2.4.2 The ArchimedeanProperty . . .

2.4.3 The Existence of 2 . . . . . . .
2.4.4 Density of Rational Numbers in R
2.4.5 Homework . . . . . . . . . . . .
Intervals . . . . . . . . . . . . . . . . . .
2.5.1 Characterization of Intervals . . .
2.5.2 Nested Intervals . . . . . . . . .
2.5.3 The Uncountability of R . . . . .
2.5.4 Homework . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3 Sequences and Series

3.1 Sequences and Their Limits . . . . . . . . . . . . . .
3.1.1 The Limit of a Sequence . . . . . . . . . . .
3.1.2 Tails of Sequences . . . . . . . . . . . . . .
3.1.3 Examples . . . . . . . . . . . . . . . . . . .
3.1.4 Homework . . . . . . . . . . . . . . . . . .
3.2 Limit Theorems . . . . . . . . . . . . . . . . . . . .
3.2.1 Examples . . . . . . . . . . . . . . . . . . .
3.2.2 Homework . . . . . . . . . . . . . . . . . .
3.3 Monotone Sequences . . . . . . . . . . . . . . . . .
3.3.1 Sequence defined recursively . . . . . . . . .
3.3.2 Limit Superior and Limit Inferior . . . . . .
3.3.3 Homework . . . . . . . . . . . . . . . . . .
3.4 Subsequences and the Bolzano-Weierstrass Theorem
3.4.1 The Existence of Monotone Subsequences .
3.4.2 The Bolzano-Weierstrass Theorem . . . . .
3.4.3 Homework . . . . . . . . . . . . . . . . . .
3.5 The Cauchy Criterion . . . . . . . . . . . . . . . . .
3.5.1 Homework . . . . . . . . . . . . . . . . . .
3.6 Properly Divergent Sequences . . . . . . . . . . . .
3.7 Introduction to Series . . . . . . . . . . . . . . . . .
3.7.1 Comparison Tests . . . . . . . . . . . . . .
4 Limits
4.1 Limits of Functions . . . . . . . . . . . . . .
4.1.1 The Definition of the Limit . . . . .
4.1.2 Sequential Criterion for Limits . . .
4.1.3 Divergence Criteria . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
22
23
23
24
24
25
26
27
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
29
29
30
31
32
32
33
37
37
38
39
40
41
42
43
44
45
46
48
49
50
50
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
51
. . . 51
. . . 52
. . . 52
. . . 53
introanal.tex; November 21, 2005; 18:02; p. 1
CONTENTS
4.2
III
4.1.4 Homework . . . . . . . . . . . . . . . . . . . . . . . . .
Limits Theorems . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Homework . . . . . . . . . . . . . . . . . . . . . . . . .
Continuous Functions
5.1 Continuous Functions . . . . . . . . . . . .
5.1.1 Homework . . . . . . . . . . . . . .
5.2 Combinations of Continuous Functions . . .
5.2.1 Composition of Continuous Functions
5.2.2 Homework . . . . . . . . . . . . . .
5.3 Continuous Functions on Intervals . . . . . .
5.3.1 The Maximum-Minimum Theorem .
5.3.2 Bisection Method . . . . . . . . . . .
5.3.3 Bolzanos Theorem . . . . . . . . .
5.3.4 Homework . . . . . . . . . . . . . .
5.4 Uniform Continuity . . . . . . . . . . . . .
5.4.1 Lipshitz Functions . . . . . . . . . .
5.4.2 The Continuous Extension Theorem
5.4.3 Homework . . . . . . . . . . . . . .
5.5 Continuity and Gauges . . . . . . . . . . . .
5.5.1 Existence of -fine Partitions . . . .
5.6 Monotone and Inverse Functions . . . . . . .
5.6.1 Inverse Functions . . . . . . . . . .
5.6.2 Homework . . . . . . . . . . . . . .
. . . .
. . . .
. . . .
. . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
Differentiation
6.1 The Derivative . . . . . . . . . . . . . . . . . . .
6.1.1 The Chain Rule . . . . . . . . . . . . . .
6.1.2 Inverse Functions . . . . . . . . . . . . .
6.1.3 Homework . . . . . . . . . . . . . . . . .
6.2 The Mean Value Theorem . . . . . . . . . . . . .
6.2.1 Inequalities . . . . . . . . . . . . . . . . .
6.2.2 Intermediate Value Property of Derivatives
6.2.3 Homework . . . . . . . . . . . . . . . . .
6.3 Taylors Theorem . . . . . . . . . . . . . . . . . .
6.3.1 Relative Extrema . . . . . . . . . . . . . .
6.3.2 Convex Functions . . . . . . . . . . . . .
6.3.3 Homework . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
53
54
56
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
57
. . . . 57
. . . . 59
. . . . 60
. . . . 61
. . . . 61
. . . . 62
. . . . 63
. . . . 64
. . . . 64
. . . . 66
. . . . 67
. . . . 68
. . . . 69
. . . . 70
. . . . 71
. . . . 71
. . . . 72
. . . . 75
. . . . 76
.
.
.
.
.
.
.
.
.
.
.
.
77
77
80
81
82
83
87
87
89
90
91
92
93
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
CONTENTS
IV
7
The Riemann Integral

7.1 The Riemann Integral . . . . . . . . . . . . . . .
7.1.1 Partitions and Tagged Partitions . . . . . .
7.1.2 Definition of Riemann Integral . . . . . .
7.1.3 Examples . . . . . . . . . . . . . . . . . .
7.1.4 Properties of the Integral . . . . . . . . . .
7.1.5 Boundedness Theorem . . . . . . . . . . .
7.1.6 Homework . . . . . . . . . . . . . . . . .
7.2 Riemann Integrable Functions . . . . . . . . . . .
7.2.1 The Squeeze Theorem . . . . . . . . . . .
7.2.2 Classes of Riemann Integrable Functions .
7.2.3 The Additivity Theorem . . . . . . . . . .
7.2.4 Homework . . . . . . . . . . . . . . . . .
7.3 The Fundamental Theorem . . . . . . . . . . . . .
7.3.1 The Fundamental Theorem (First Form) .
7.3.2 The Fundamental Theorem (Second Form)
7.3.3 Substitution Theorem . . . . . . . . . . .
7.3.4 Lebesgues Integrability Criterion . . . . .
7.3.5 Integration by Parts . . . . . . . . . . . .
7.3.6 Homework . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
95
95
95
96
97
97
98
98
99
100
101
102
103
104
104
105
107
107
110
111
Topology of Real Numbers

8.1 Open and Closed Sets in R . . . . . . . . .
8.1.1 The Characterization of Closed Sets
8.1.2 The Characterization of Open Sets
8.2 Compact Sets . . . . . . . . . . . . . . . .
8.3 Continuous Functions . . . . . . . . . . .
8.3.1 Continuity . . . . . . . . . . . . .
8.3.2 Preservation of Compactness . . .
8.4 Metric Spaces . . . . . . . . . . . . . . . .
8.4.1 Metrics . . . . . . . . . . . . . . .
8.4.2 Neighborhoods and Convergence .
8.4.3 Cauchy Sequences . . . . . . . . .
8.4.4 Open Sets and Continuity . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
113
113
115
115
115
116
116
116
116
116
117
117
118
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Bibliography
118
Notation
121
Chapter 1
Preliminaries
1.1
Sets and Functions
Reading: Appendix A (Logic and Proofs).

Set.
Elements (members).
Subset.
Proper subset.
Definition 1.1.1 Two sets A and B are equal if they contain the same
elements.
Alternatively, A = B if A B and B A.
Description of sets.
Empty set.
Universal set.
Examples.
1. Natural numbers N
2. Integer numbers Z
3. Rational numbers Q
1
CHAPTER 1. PRELIMINARIES
2
4. Real numbers R
5. Complex numbers C
1.1.1
Set Operations
Definition 1.1.2
The union of the sets A and B is the set

A B = {x | x A or x B}
The intersection of the sets A and B is the set

A B = {x | x A and x B}
The complement of the set B relative to the set A is the set

A \ B = {x | x A and x < B}
The sets A and B are disjoint if they have no common elements, that is
AB=
Theorem 1.1.1 De Morgans Laws
A \ (B C) = (A \ B) (A \ C)
A \ (B C) = (A \ B) (A \ C)
Proof: Diagrams.
Infinite collection of sets.
{Ak }
k=1
Union of a collection of sets
[
Ak = {x | x Ak for some k N}
k=1
Intersection of a collection of sets
\
Ak = {x | x Ak for all k N}
k=1
Examples.
1.1. SETS AND FUNCTIONS
1.1.2
Cartesian Products
Definition 1.1.3 The Cartesian product of two nonempty sets A and
B is the set of all ordered pairs (a, b) with a A and b B, that is
A B = {(a, b) | a A, b B}
Example.
1.1.3
Mappings and Functions
Function as a mapping.
f : A B,
f
a 7 b,
f maps A into B
image of a under f
b = f (a),
value of f at a
Definition 1.1.4
A function from a set A to a set B is a rule that
assigns to each element of the set A an element of the set B.
The set A is the domain of the function f , denoted by D( f ), and the set
B is the codomain of the function f .
The range of the function f is the subset R( f ) of the codomain B such
that for each b R( f ) there exists a A such that b = f (a).
Remark. The range of a function is a subset of the codomain:

D( f ) = A,
R( f ) B
Diagrams.
Function as a relation.
Definition 1.1.5 A function from a set A to a set B is a set f of ordered

pairs (a, b) in A B such that for each a A there exists a unique b B
with (a, b) f . That is, if (a, b) f and (a, b0 ) f , then b = b0 .
1.1.4
Direct and Inverse Images

Definition 1.1.6 Let f : A B be a function. The direct image of a
subset E A under f is the subset of B
f (E) = { f (x) | x E}
The inverse image of a subset H B under f is the subset of A

f 1 (H) = {x A | f (x) H} .
Remark. The range of a function is the image of the domain:
R( f ) = f (A)
Examples. Diagrams.
1.1.5
Special Types of Functions

Definition 1.1.7
Let f : A B be a function.
f is injective (one-to-one) if whenever x1 , x2 then f (x1 ) , f (x2 ).
f is surjective (maps A onto B) if f (A) = B, or R( f ) = B.

That is, if for any b B there exists a A such that f (a) = b.
f is bijective if it is both injective and surjective (one-to-one and onto).
Example.
1.1.6
Inverse Functions
Definition 1.1.8
Let f : A B be a bijection of A onto B. The
inverse function f 1 : B A is defined by
f 1 = {(b, a) B A | (a, b) f }
That is, for any b B, f 1 (b) = a where a A is such that f (a) = b.
Remarks.
D( f ) = R( f 1 ),
R( f ) = D( f 1 ) .
1.1. SETS AND FUNCTIONS
1.1.7
Composition of Functions
Definition 1.1.9 Let f : A B and g : B C be two functions.

Then the composite function g f : A C is the function from A into
C defined by
(g f )(x) = g( f (x)),
x A .
Remarks.
g f , f g
D(g f ) = D( f )
Examples
Theorem 1.1.2 Let f : A B and g : B C. If H C, then
(g f )1 (H) = f 1 (g1 (H)) .

Proof: Exercise!
1.1.8
Restrictions of Functions
Let f : A B and A1 A. The restriction of f to A1 is the function

f1 : A1 B defined by
f1 (x) = f (x),
x A1 .
Notation
f1 = f |A1
1.2
Finite and Infinite Sets
Reading: Proofs are in Appendix B.

Definition 1.2.1
Empty set has 0 elements.
A set S has n elements if there is a bijection f : Nn S from Nn =

{1, 2, . . . , n} onto S .
A set S is finite if it is either empty or has n elements for some n N.
A set S is infinite if it is not finite.
Theorem 1.2.1 Uniqueness Theorem. The number of elements of a

finite set is unique.
Theorem 1.2.2 The set N is infinite.

Theorem 1.2.3
1. Let A and B be disjoint sets. If A has m elements
and B has n elements, then the set A B has n + m elements.
2. Let C A. If A has m elements and C has 1 element, then A \ C

has m 1 elements.
3. If A is an infinite set and B is a finite set, then the set A \ B is
infinite.
Theorem 1.2.4 Let B A. Then:
1. If A is finite, then B is finite.

2. If B is infinite, then A is infinite.
1.2.1
Countable Sets
Definition 1.2.2 A set S is denumerable (countably infinite) if there
is a bijection f : N S .
A set S is countable if it is either finite or countably infinite.

A set S is uncountable if it is not countable.
1.2. FINITE AND INFINITE SETS
Examples.
Remark. The union of countably infinite sets is countably infinite.
Theorem 1.2.5 The set N N is countably infinite.

Proof: Diagonal argument.
Theorem 1.2.6 Let B A. Then:
1. If A is countable, then B is countable.

2. If B is uncountable, then A is uncountable.
Theorem 1.2.7 The following statements are equivalent:
1. S is a countable set.
2. There is a surjection g : N S .
3. There is an injection f : S N.
Theorem 1.2.8 The set Q is countably infinite.

Proof:
1. Surjection g : N N Q+ .
2. Bijection f : N N N.
3. Surjection g f : N Q+ .
1.2.2
Theorem 1.2.9 The union of a countable collection of countable sets is

countable.
Proof: Diagonal argument.
Theorem 1.2.10 Cantors Theorem. Let A be a set and P(A) be the
set of all subsets of A. There is no surjection f : A P(A).
Homework
Reading: Appendix A, B. Sec. 1.1-1.3.

Exercises: 1.1[19], 1.3[1,5,7,9].
Chapter 2
The Real Numbers
2.1
The Algebraic and Order Properties of R
R is a field with respect to addition and multiplication.

Binary operation
B : R R R,
(a, b) 7 B(a, b)
Definition 2.1.1 Algebraic properties of R. There are two binary

operations on R: addition, +, and multiplication, , satisfying the following properties: a, b, c R
1. a + b = b + a
2. (a + b) + c = a + (b + c)
3. 0 R such that
0+a=a
4. a R (a) R such that
a + (a) = 0
5. ab = ba
9
CHAPTER 2. THE REAL NUMBERS
10
Definition 2.1.2
1. (ab)c = a(bc)
2. 1 R, such that 1 , 0 and

1a=a
3. a R, a , 0, a1 R such that
aa1 = 1
4. a(b + c) = ab + ac
Theorem 2.1.1
1. Let x, a R and x + a = a. Then x = 0.
2. Let y, b R and yb = b. Then y = 1.

3. For any a R, 0a = 0.
Proof:
1. x = x + 0 = x + (a + (a)) = (x + a) + (a) = a + (a) = 0.
2. 0a + a = 0a + 1a = (0 + 1)a = 1a = a. So 0a = 0.
Theorem 2.1.2
1. Let a, b R, a , 0 and ab = 1. Then b = a1 .
2. Let a, b R and ab = 0. Then either a = 0 or b = 0.

Proof:
1. b = 1b = (aa1 )b = a1 (ab) = a1
2. If a , 0 and b , 0, then 1 = b1 a1 ab = b1 a1 0 = 0. Contradiction.

Subtraction is defined by: a, b R
a b = a + (b)
2.1. THE ALGEBRAIC AND ORDER PROPERTIES OF R
11
Division is defined by: a, b R, b , 0,

a
= ab1
b
2.1.1
Rational and Irrational Numbers
Identify N = Z+ with the subset of R by

n=1
+}
1
| + {z
n
Define Z as a subset of R as follows. Identify the zeros

0Z = 0R
Define the negative integers as
n = (1) + + (1)
|
{z
}
n
or as the opposite to n. Then

Z = Z+ {0} Z
Define the rational numbers Q as the subset of R by

a
Q = r = a, b Z, b , 0
b
The set of irrational numbers is R \ Q
Theorem 2.1.3 There does not exist a rational number r Q such that
r2 = 2.
Proof: By contradiction.
1. Let r = p/q and r2 = 2.
2. Then p2 = 2q2 .
3. So, p is even. So, p = 2m for some m N.
4. Thus, q is odd.
5. q2 = 2m2 . So, q is even. Contradiction.

12
2.1.2
The order Properties of R

Definition 2.1.3 The Order Properties of R. There is a nonempty
subset R+ R, the set of positive real numbers, that satisfies the properties: a, b R+
1. a + b R+ ,
2. ab R+ ,
3. a R+ exactly one of the following holds:
a R+ ,
(a) R+ ,
a = 0.
The set of negative real numbers

R = {a | a R+ }
Trichotomy Property. R is a disjoint union of the following sets
R = R {0} R+
that is
0 < R+ ,
0 < R ,
R R+ =
Positive real numbers

a R+ ,
a>0
Non-negative real numbers

a R+ {0},
a0
Negative real numbers

a R ,
a<0
Non-positive real numbers

a R {0},
Definition 2.1.4
a 0.
Let a, b R.
1. If a b > 0, then a > b (or b < a).

2. If a b 0, then a b (or b a)
2.1. THE ALGEBRAIC AND ORDER PROPERTIES OF R
13
Trichotomy Property. For any a, b R exactly one of the following is true

a > b,
a = b,
a<b
Corollary.
If a b and a b, then a = b.
Theorem 2.1.4 Let a, b, c R.
1. If a > b and b > c, then a > c.
2. If a > b, then a + c > b + c.

3. If a > b and c > 0, then ac > bc.
4. If a > b and c < 0, then ac < bc.
Proof:
1. Since a b > 0 and b c > 0, then (a b) + (b c) > 0. Thus, a c > 0
and a > c.

Theorem 2.1.5
1. Let a R, a , 0. Then a2 > 0.
2. 1 > 0.
3. For any n N, n > 0.
Proof:
1. We have a > 0 or a < 0. If a > 0, then a2 > 0. If a < 0, then (a) > 0
and a2 = (a)(a) > 0.
2. 1 = 12 . So 1 > 0.
3. n = 1 + + 1. So, n > 0.
Remark. There is no smallest positive real number.
Theorem 2.1.6 Let a R. If for any > 0, 0 a < , then a = 0.

Proof: By contradiction.
14
1. Suppose a > 0. Let = a/2. Get contradiction.

Theorem 2.1.7 Let a, b R and ab > 0. Then either
1. a > 0 and b > 0, or

2. a < 0 and b < 0.
Proof:
1. We have a , 0 and b , 0.
2. If a > 0, then a1 > 0.
3. Then b = a1 (ab) > 0.

Corollary 2.1.1 Let a, b R and ab < 0. Then either
1. a > 0 and b < 0, or

2. a < 0 and b > 0.
2.1.3
Inequalities
Examples.
2.2. ABSOLUTE VALUE AND REAL LINE
2.2
15
Absolute Value and Real Line

Definition 2.2.1
The absolute value of a real number a is defined by
a, if a > 0
0, if a = 0
|a| =
a, if a < 0
Theorem 2.2.1 Properties the Absolute Value. Let a, b, c R and

c 0. Then
1. |ab| = |a||b|
2. |a|2 = a2
3. |a| c if and only if c a c
4. a |a| |a|
Proof: Exercise.
Theorem 2.2.2 Triangle Inequality. Let a, b R. Then
|a + b| |a| + |b|
Proof: Easy.
Corollary 2.2.1 Let a, b R. Then
1. ||a| |b|| |a b|,

2. |a b| |a| + |b|
Corollary 2.2.2 Let a1 , . . . , an R. Then

|a1 + + an | |a1 | + + |an |
Examples.
16
2.2.1
The Real Line

Definition 2.2.2
the set
Let a, R and > 0. The -neighborhood of a is

B (a) = {x R | |x a| < }
Theorem 2.2.3 Let a R. If for any > 0, x B (a), then x = a.

Proof:
1. We have |x a| < . So |x a| = 0.
Examples.
2.2.2
Homework
Exercises: 2.1[13,20,23], 2.2[14,15]
2.3. THE COMPLETENESS PROPERTY OF R
2.3
The Completeness Property of R
2.3.1
Suprema and Infima
17
R is a complete ordered field.

Definition 2.3.1
Let S R be a nonempty set of real numbers.
1. S is bounded above if there exists u R such that s S , s u.

Such u is an upper bound of S .
2. S is bounded below if there exists v R such that s S , s v.

Such v is a lower bound of S .
3. S is bounded if it is bounded above and below.
That is there is M R such that s S , |s| M.
4. S is unbounded if it is not bounded.
That is for any M R there is s S such that |s| > M.
Example.
Definition 2.3.2
Let S R be a nonempty set of real numbers.
1. Let S be bounded above. Then u R is a supremum (least

upper bound) of S if:
(a) u is an upper bound of S , and
(b) if v is any upper bound of S , then u v.

2. Let S be bounded below. Then t R is an infimum (greatest
lower bound) of S if:
(a) t is a lower bound of S , and
(b) if v is any lower bound of S , then t v.
The supremum and infimum, when they exist, are unique.

Proof: Exercise.
Notation.
sup S ,
inf S
18
Not every set S R has the supremum (or infimum).

For a set to have the supremum it must be: 1) nonempty and 2) bounded
above.
For a set to have the infimum it must be: 1) nonempty and 2) bounded
below.
For a set to have both the supremum and infimum it must be: 1) nonempty
and 2) bounded.
Equivalent statements about an upper bound:
1. u is an upper bound of S
2. s S , s u.
Let u = sup S . Then the following are equivalent statements:
1. if v is any upper bound of S , then u v,
2. if w < u, then w is not an upper bound of S ,
3. if w < u, then s S such that s > w,
4. for any > 0, s S such that u < s.
Alternative formulations for the supremum:
Lemma 2.3.1 Let S R be a nonempty set of real numbers. Then
u = sup S if and only if
1. s S , s u, and
2. if v < u, then there is s0 S such that v < s0 .
Proof: Exercise.
Lemma 2.3.2 Let S R be a nonempty set of real numbers and u R

be an upper bound of S . Then u = sup S if and only if > 0 there is
s S such that u < s.
Proof:
1.

2.3. THE COMPLETENESS PROPERTY OF R
19
Remarks.
Supremum of a set does not have to belong to that set.
If u = sup S and u S , then u = max S is the largest element of the set S .
If S is finite, then sup S is its largest element, and inf S its least element.
Examples.
2.3.2
The Completeness Property of R
The final axiom about R.
Definition 2.3.3 The Completeness Property of R. Every nonempty

set of real numbers bounded above has a supremum in R.
Corollary. Every nonempty set of real numbers bounded below has an

infimum in R.
Proof: Exercise.
2.3.3
Homework
Exercises: 2.3[6,7,8,9,10]
20
2.4
Applications of Supremum Property
Examples.
Compatibility of sup and inf with algebraic properties of R. Let S R,
S , . Let a R and a + S be the set
a + S = {a + s|s S }
If S is bounded above, then a + S is bounded above and
sup(a + S ) = a + sup S .
Proof: Exercise.
Proposition. Let A, B R, A, B , . If a A, b B, a b, then
sup A inf B .
Proof: Exercise.
2.4.1
Functions
Definition 2.4.1 Let f : D R.
1. f is bounded above if f (D) is bounded above.

2. f is bounded below if f (D) is bounded below.
3. f is bounded if f (D) is bounded.
Example. Let f, g : D R be bounded and x D, f (x) g(x). Then

sup f (D) sup g(D)
Proof: Easy.
Note: there is no relation between sup f and inf g.
Example. Let f, g : D R be bounded and x, y D, f (x) g(y). Then
sup f (D) inf g(D)
Proof: Easy.
2.4. APPLICATIONS OF SUPREMUM PROPERTY
2.4.2
21
The Archimedean Property

Theorem 2.4.1 For any x R, there exists n N such that x < n.
That is N is not bounded above in R.
Proof:
1. By contradiction. Suppose x R such that n N, n x.
2. Then x is an upper bound of N. So, a = sup N.
3. Then a 1 is not an upper bound of N.
4. m N such that m > a 1, or a < m + 1.
5. Thus, a is not an upper bound.
Corollary 2.4.1 Let S =
n
o
1
n N . Then inf S = 0.
n
Proof:
1. Let b = inf S . We have b 0.
2. > 0, n N, such that 0 b
1
n
< .
3. b = 0.
4. Or, by contradiction. Suppose b > 0. Get a contradiction.

Corollary 2.4.2 For any t R, t > 0, there exists n N such that
0<
1
<t
n
Proof:
1. By Archimedean property.

Corollary 2.4.3 For any y R, y > 0, there is n N such that
n 1 y < n.
Proof:
22
1. Let E = {m N|y < m}
2. By Well Ordering Principle least element of E, say n.

3. Then n 1 < E.
4. So, n 1 y < n.
2.4.3
The Existence of
Theorem 2.4.2 There exists x R+ such that x2 = 2.

Proof:
1. Let S = {s R|0 s < 2}.
2. S , and S is bounded above.
3. x = sup S .
4. We have x > 1.
5. Consider x2 . By Trichotomy Property x2 < 2 or x2 > 2 or x2 = 2.
6. Case x2 < 2. This contradicts the fact that x is an upper bound of S .

2
Since n N such that x + 1n < 2 and thus x + 1/n S . Or
x + 1/n > x.
7. Case x2 > 2. This contradicts the fact that x is the least upper bound
of S .
Since m N such that x 1/m is also an upper bound of S .
Since (x 1/m)2 > 2 and s2 < 2 < (x 1/m)2 . So s < x 1/m.
8. Thus x2 = 2.
Corollary 2.4.4 For any a R+ there is a unique b > 0 such that

b2 = a.
b is the positive square root of a, b = a.
Corollary 2.4.5 For any a R+ and any n N there is a unique b > 0

such that bn = a.
b is the positive n-th root of a, b = a1/n .

2.4. APPLICATIONS OF SUPREMUM PROPERTY
2.4.4
23
Density of Rational Numbers in R

Theorem 2.4.3 The Density Theorem. Let x, y R and x < y. Then
there is r Q such that x < r < y.
Proof:
1. Case 0 < x < y.
2. Case x < y < 0.
3. Case x < 0 < y.
Corollary 2.4.6 Let x, y R and x < y. Then there is an irrational

number z such that x < z < y.
Proof:
1.
2.4.5
Homework
Exercises: 2.4[7,9,12], 2.5[3,6,10]
24
2.5
Intervals
Open interval
Endpoints of an interval
Closed interval
Half-open (half-closed) intervals
Length of an interval
Infinite open intervals
Infinite closed intervals
Remark. R and are the only sets that are both open and closed.
2.5.1
Characterization of Intervals
Theorem 2.5.1 Characterization Theorem. If S R contains at least
two points and
for any x, y S , x < y implies [x, y] S ,

then S is an interval.
Proof: Four cases.
1. Case I. S is bounded. Let a = inf S and b = sup S .
2. Then S [a, b].
3. Claim: (a, b) S .
4. Thus, S is one of the bounded intervals.
5. Case II. S is bounded above and unbounded below.
6. Let b = sup S .
7. Then S (, b].
8. Claim: (, b) S .
9. Thus, S is one of the two intervals unbounded below.
10. Case III. S is bounded below and unbounded above.
2.5. INTERVALS
25
11. Similarly to II.

12. Case IV. S is unbounded above and below.
13. Claim: S = R.
2.5.2
Nested Intervals
Definition 2.5.1 A sequence of intervals {In }
n=1 is nested if
I1 I2 In In+1
Examples.
Theorem 2.5.2 Nested Intervals Property. The intersection of a
nested sequence of closed bounded intervals is nonempty.
or
If {In }
n=1 is a nested sequence of closed bounded intervals, then there
exists R, such that In for each n N.
Proof:
1. Completeness of R is essential for this!
2. Let In = [an , bn ].
3. Then
an an+1 b1 and bn bn+1 a1
4. Let S = {an |n N}.
5. Let = sup S .
6. Then
an bn ,
n N

26
Theorem 2.5.3 If {In }

n=1 is a nested sequence of closed bounded intervals, In = [an , bn ], such that
inf{(bn an ) | n N} = 0,
then there exists a unique real number R, such that In for each
n N.
Proof:
1. Let
= inf{bn |n N} and = sup{an |n N}.
2. Then
an bn for any n N.
3. Next, > 0, m N such that n m,
bn an < .
4. We have
0 bn an < .
5. Thus = .
2.5.3
The Uncountability of R
Theorem 2.5.4 The set R is uncountable.
Proof:
1. Consider I = [0, 1] R.
2. Claim: I is uncountable.
3. Assume I = {x1 , x2 , . . . } is countable.
4. Let I1 be a closed bounded interval such that I1 I and x1 < I1 .
5. Let In be a closed bounded interval such that In In1 and xn < In .
6. Then In for each n N.
7. Thus, xn , for each n N.
2.5. INTERVALS
27
8. Contradiction.

Remark. The set of irrational numbers R \ Q is uncountable.
2.5.4
Homework
Exercises: 2.5[3,6,10]
28
Chapter 3
Sequences and Series
3.1
Sequences and Their Limits
Sequence (an ), an S , or f : N S .
Definition 3.1.1
R.
A sequence of real numbers is a a function X : N
Elements (terms) of a sequence xn = X(n).

Notation X = (xn ) = (xn |n N) = (xn )nN = (xn )
n=1
Remark. The set {xn |n N} (non-ordered collection of elements) is different from the sequence (xn |n N) (an ordered list of elements).
Description of sequences:
1. list,
2. formula,
3. recursion.
Examples.
Definition 3.1.2
n N.
A sequence X = (xn ) is constant if xn = x for each
29
CHAPTER 3. SEQUENCES AND SERIES
30
3.1.1
The Limit of a Sequence

Definition 3.1.3 Convergence, Limit, Divergence. A sequence X =
(xn ) converges to x R (or has a limit x) if > 0, K N such that
n K,
|xn x| < .
The number x is the limit of the sequence X.

If X has a limit, then X converges, or is convergent.
If X does not have a limit, then X diverges, or is divergent.
Notation. lim X = x,
lim xn = x,
xn x
Theorem 3.1.1 Uniqueness of Limit. A sequence can have at most

one limit.
Proof:
1. Suppose X has two limits x and y.
2. Show that > 0, |x y| < .
3. So, x = y.

Theorem 3.1.2 Let X = (xn ) be a sequence in R and x R. The following statements are equivalent:
1. lim X = x,
2. > 0, K N such that n K,
|xn x| < ,
3. > 0, K N such that n K,
x < xn < x + ,
4. for any -neighborhood B (x) of x, K N such that n K,
xn B (x).
3.1. SEQUENCES AND THEIR LIMITS
31
Proof: Follows from definitions.

Corollary 3.1.1 Let X = (xn ) be a sequence in R and x R. Then

lim X = x if for any -neighborhood B (x) of x all but a finite number
of terms of X belong to B (x).
Examples.
Remark. To show that X = (xn ) does not converge to x R it is enough to
find 0 > 0 such that K N, n K such that
|xn x| .
Theorem 3.1.3 Let X = (xn ) be a sequence of real numbers, x R

be a real number, and A = (an ) be a sequence of positive real numbers
converging to zero, that is lim A = 0. Suppose that there exists a positive
real number C R+ and some integer M N such that n M,
|xn x| Can .
Then lim X = x.
Proof:
1. Easy.
3.1.2
Tails of Sequences
Definition 3.1.4 Let X = (xn )

n=1 be a sequence in R and M N. The
M-tail of the sequence X is the sequence X M = (xn )

n=M+1 = (x M+k )k=1 .
Theorem 3.1.4 Let X = (xn ) be a sequence in R and m N. Then the

m-tail Xm converges if and only if X converges and lim Xm = lim X.
Proof:
1. Let X M = Y = (yn ) = (xn+M ).
2. I. Assume X x. Show Y x.
3. II. Assume Y x. Show X x.

32
3.1.3
Examples
3.1.4
Homework
Exercises: 3.1[3(b),8,10,13,14]
3.2. LIMIT THEOREMS
3.2
33
Limit Theorems
Definition 3.2.1 Let X = (xn ), Y = (yn ) and Z = (zn ) be sequences in

R, c R and f : R R be a function.
The multiple of X by c is the sequence
cX = (cxn )
We say that
X c if xn c for all n N.
We say that
X , 0 if xn , 0 for all n N.
The absolute value of the sequence X is the sequence
|X| = (|xn |)
We say that
X Y if xn yn for all n N.
The function of a sequence X is the sequence
f (X) = ( f (xn ))
The sum and the difference of X and Y are the sequences
X + Y = (xn + yn ) and X Y = (xn yn )
The product of X and Y is the sequence
XY = (xn yn )
If Z , 0, then the quotient of X and Z is the sequence
!
xn
X/Z =
zn
34
Definition 3.2.2 A sequence X = (xn ) in R is bounded if there exists

M R such that |X| M,
That is n N
|xn | M.
Theorem 3.2.1 Any convergent sequence in R is bounded.

Proof: Easy.
Lemma 3.2.1 Let Z be a sequence in R such that Z , 0 and lim Z , 0.

Then 1/Z is a bounded sequence.
Proof:
1. Get an estimate (for large n)
1
3
|z| < |zn | < |z|
2
2
2. Then
1
2
<
|zn | |z|
Theorem 3.2.2 Let X and Y be convergent sequences in R, and Z be a

convergent sequence such that Z , 0 and lim Z , 0. Let c R. Then
X + Y, X Y, XY, cX, X/Z converge and
lim X Y = lim X lim Y
lim XY = lim X lim Y

lim cX = c lim X
lim X/Z =
lim X
lim Z
Proof:
1. For X Y use triangle inequality.
3.2. LIMIT THEOREMS
35
2. For XY use
|xn yn xy| |xn ||yn y| + |y||xn x|
3. For X/Z use

xn x |xn x| + |x| |z zn |
zn z
|zn |
|z| |zn |
4. We have
1
2
<
|zn | |z|
5. Then for large n

xn x 2 |x x| + 2|x| |z z |
n
zn z |z| n
|z|2
Theorem 3.2.3 Let X be a convergent sequence in R.

If X 0, then lim X 0.
Proof:
1. By contradiction. Suppose that x = lim X < 0.
2. Let < |x| and get a contradiction.
Theorem 3.2.4 Let X and Y be convergent sequences in R.

If X Y, then lim X lim Y.
Proof:
1. Let Z = Y X.
2. Then lim Z 0.
Theorem 3.2.5 Let X be a convergent sequence in R and a, b R.

If a X b, then a lim X b.
Proof:
1. Let A = (a) and B = (b).
36
2. Follows from previous theorem.

Theorem 3.2.6 Squeeze Theorem. Let X and Z be convergent sequences in R and lim X = lim Z = x R. Let Y be a sequence in
R.
If X Y Z, then Y converges and lim Y = x.
Proof: Easy.
Examples.
Theorem 3.2.7 Let X be a convergent sequence in R. Then the sequence |X| converges and
lim |X| = | lim X|.
Proof:
1. By Triangle inequality.
Theorem 3.2.8 Let X be a convergent sequence in R and X 0.
The sequence X converges and lim X = lim X.

Proof:
1. Case I. x = lim X = 0.
2. Case II. x > 0.
3. We estimate
1
| xn x| |xn x|.
x
Theorem 3.2.9 Let X = (xn ) be a convergent sequence in R and X > 0.

Let Y = (yn ) be a sequence defined by yn = xn+1 /xn . If Y converges and
lim Y < 1, then X converges and lim X = 0.
Proof:
1. We have Y > 0.
3.2. LIMIT THEOREMS
37
2. Therefore, L = lim Y 0.
3. Assume L < 1. Let 0 L < r < 1.
4. Let = r L.
5. For large n,

xn+1 L <

xn
6. So,
0 < xn+1 < xn r
7. Then there is a C such that for large n
0 < xn+1 < Crn+1
8. Thus lim xn = 0.
3.2.1
Examples
3.2.2
Homework
Exercises: 3.2[3,5(b),7,9,17]
38
3.3
Monotone Sequences
Definition 3.3.1
Let X = (xn ) be a sequence in R.
X is increasing if
xn xn+1 for any n N.
X is decreasing if
xn xn+1 for any n N.

X is monotone if it is either increasing or decreasing.
Examples.
Notation. Let X = (xn ) be a sequence in R defined by a function X : N R.
The set of all elements of the sequence is the range of this function
X(N) = {xn |n N}.
Theorem 3.3.1 Monotone Convergence Theorem. Let X be a monotone sequence in R.
X is convergent if and only if it is bounded.
(i) If X is bounded and increasing, then lim X = sup X(N).

(ii) If X is bounded and decreasing, then lim X = inf X(N).
Proof:
1. Claim: Convergence implies boundedness. Proved before.
2. Claim: Boundedness and monotonicity implies convergence.
3. Let X be increasing and bounded.
4. Then the set X(N) is bounded.
5. Let b = sup X(N).
6. Claim: b = lim X.
7. Let > 0.
8. Then b is not an upper bound of X(N).
3.3. MONOTONE SEQUENCES
39
9. So, there exists k N such that xk > b .

10. Hence, n k, xn > b .
11. Thus, lim X = b.
12. Similarly, if X is decreasing.
3.3.1
Sequence defined recursively

Proposition 3.3.1 (Square Roots.) Let a > 0 and X = (xn ) be a sequence defined by
!
1
a
x1 = 1,
xn+1 =
xn +
2
xn
Then X converges and
lim X =
a.
Proof:
1. We have
2
xn2 2xn+1 xn + a = (xn xn+1 )2 xn+1
+a=0
2. Therefore, xn2 a.
3. Next, we have
xn xn+1
!
x2 a
a
1
= xn
xn +
= n
0
2
xn
2xn
4. Therefore,
a xn+1 xn
5. Compute the limit from

1
a
x=
x+
2
x

40
Example. Let
1
xn
Show, X is bounded above and increasing, therefore, converging.
x1 = 1,
xn+1 = 2 +
Compute the limit from

x=2+
3.3.2
1
x
Limit Superior and Limit Inferior

Definition 3.3.2 Let X be a bounded sequence in R and Xn be the n-tail
of the sequence X. Let Xn (N) be the set of elements of the tail Xn
Xn = {xn+k |k N}
Let A and B be the sequences defined by
an = inf Xn (N),
bn = sup Xn (N).
The limit superior of the sequence X is the limit of the sequence B, that
is
lim sup X = lim sup Xn (N)
n
The limit inferior of the sequence X is the limit of the sequence A, that
is
lim sup X = lim inf Xn (N)
n
Then the sequences A and B are bounded and monotone, A is increasing and
B is decreasing.
Therefore, the sequences A and B are convergent.
Notation.
lim sup X = limX,
lim inf X = limX.
Definition 3.3.3 Let X be a sequence in R. A real number a R is

a limit point of the sequence X if every neighborhood of a contains
infinitely many elements of the sequence X.
A limit point of a sequence does not have to be an element of the sequence.

3.3. MONOTONE SEQUENCES
41
Every point that appears infinitely many times in a sequence is a limit point
of the sequence.
Proposition 3.3.2 Let X be a bounded sequence in R and P be the set

of all limit points of the sequence X. Then
lim sup X = sup P,
and
lim inf X = inf P
A sequence X converges if and only if

lim sup X = lim inf X
In this case
lim X = lim sup X = lim inf X.
3.3.3
Homework
Exercises: 3.3[1,4,5,7,9,10]
42
3.4
Subsequences and the Bolzano-Weierstrass Theorem

Lemma 3.4.1 Let A be a strictly increasing function A : N N such
that
A(k + 1) > A(k),
k N.
Then
A(k) k,
k N.
Proof: Obvious.
Example.
Definition 3.4.1 Let X = (xn ) be a sequence in R and A = (nk ) be a

strictly increasing sequence in N. A subsequence of X is the sequence
X 0 = (xA(k) ) = (xnk ) = (xn1 , xn2 , . . . ) = (xnk )
k=1 = (xnk )kN .
A subsequence X 0 : N R is the composition of two sequences A : N N

and X : N R, that is
X 0 = X A,
or
X 0 (k) = X(A(k)), k N
A tail of a sequence is a subsequence.

Example.
Theorem 3.4.1 Let X be a sequence in R and X 0 be a subsequence of

X. If X converges, then X 0 also converges and
lim X 0 = lim X
Proof: Easy.
Example.
3.4. SUBSEQUENCES AND THE BOLZANO-WEIERSTRASS THEOREM43

Theorem 3.4.2 Let X be a sequence in R and a R. The following are
equivalent statements:
1. X does not converge to a.
2. there exists 0 such that k N, nk N such that nk k and

|xnk a| 0 .
3. there exists 0 and a subsequence X 0 = (xnk ) of X such that k
N,
|xnk a| 0 .
Proof:
1. (1) implies (2). Use negation of the definition of lim X = a.
2. (2) implies (3). Construct a subsequence X 0 from (2).
3. (3) implies (1). By contradiction.
Proposition 3.4.1 (Divergence Criteria.) Let X be a sequence in R. If

X is: either (i) unbounded or (ii) X has two convergent subsequences
X 0 and X 00 such that lim X 0 , lim X 00 , then X diverges.
Proof: Exercise.
Examples.
3.4.1
The Existence of Monotone Subsequences

Theorem 3.4.3 Monotone Subsequence Theorem. Any sequence in
R has a monotone subsequence.
Proof:
1. We say that xm is a peak if n m,
xm xn
That is xm is an upper bound for the m-tail Xm , or
Xm xm .
44
2. Case I. X has infinitely many peaks.

3. Then the sequence of peaks is a decreasing subsequence of X.
4. Case II. X has finitely many peaks.
5. Then the tail beyond the last peak does not have any peaks.
6. Since all the elements in the tail are not peaks, construct an increasing
subsequence of X.
3.4.2
The Bolzano-Weierstrass Theorem

Theorem 3.4.4 The Bolzano-Weierstrass Theorem. A bounded sequence in R has a convergent subsequence.
Proof:
1. (I) Indirect. X has a monotone subsequence X 0 .
2. X 0 is bounded and monotone, hence, convergent.
3. (II) Constructive.
4. Construct a sequence of closed bounded intervals Ik = [ak , bk ] such
that Ik Ik+1 with the lengths bk ak = 2k (b1 a1 ) with the property
that
X(N) Ik ,
5. Construct a subsequence X 0 such that xnk Ik .
6. Then there exists a unique element Ik , k N.
7. So,
|xnk | < 2k (b1 a1 )
8. Thus, lim X = .
Theorem 3.4.5 Let X be a bounded sequence in R and a R. If every convergent subsequence of X converges to a, then the sequence X
converges to a.
Proof:
3.4. SUBSEQUENCES AND THE BOLZANO-WEIERSTRASS THEOREM45

1. By contradiction. Suppose that X does not converge to a.
2. Then there exists 0 > 0 and a subsequence X 0 such that k N,
|xnk a| 0 .
3. Since X 0 is bounded, X 0 has a convergent subsequence X 00 .
4. Then X 00 converges to a.
5. This contradicts the above.
3.4.3
Homework
Exercises: 3.4[2,3,5,6,9]
46
3.5
The Cauchy Criterion

Definition 3.5.1 A sequence X = (xn ) in R is a Cauchy sequence if
> 0, K N such that n, m K,
|xn xm | < .
Examples.
Lemma 3.5.1 Every convergent sequence is Cauchy.

Proof: Easy.
Lemma 3.5.2 Every Cauchy sequence is bounded.

Proof: Easy.
Theorem 3.5.1 Cauchy Convergence Criterion. A sequence of real

numbers is convergent if and only if it is Cauchy.
Proof:
1.
2.
3.
4.
5.
6.
7.
(I). Proved above.

(II). Let X be Cauchy.
Then X is bounded.
So, there exists a convergent subsequence X 0 .
Claim: lim X = lim X 0 .
Use the fact that X is Cauchy and X 0 converges to a = lim X 0 .
Get an estimate for some K N
|xn a| |xn xK | + |xK a| <
Examples.
Definition 3.5.2
A sequence X is contractive if C R such that
0 < C < 1 and n N
|xn+2 xn+1 | C|xn+1 xn |.

The number C is the constant of the contractive sequence.
3.5. THE CAUCHY CRITERION
47
Theorem 3.5.2 Every contractive sequence is Cauchy and, therefore,

convergent.
Proof:
1. We have
|xn+2 xn+1 | C n |x2 x1 |
2. Let m > n. Then
|xm xn | |xm xm1 | + + |xn+1 xn |
(C m2 + + C n1 )|x2 x1 |
mn
n1 1 C
C
|x2 x1 |
1C
1
C n1
|x2 x1 |
1C
3. So, X is Cauchy.

Corollary 3.5.1 Let X = (xn ) be a contractive sequence with constant
C, 0 < C < 1, and a = lim X. Then n N
1.
|a xn |
2.
|a xn+1 |
C n1
|x2 x1 |,
1C
C
|xn+1 xn |.
1C
Proof:
1. (1). We have m > n
|xm xn | C n1
1
|x2 x1 |
1C
2. As m , we get (1).
3. (2). We have for m > n
|xm xn | |xm xm1 | + + |xn+1 xn |
48
4. By induction
|xn+k xn+k1 | C k |xn xn1 |

5. Thus
|xm xn | (C mn + + C)|xn xn1 |
C
|xn xn1 |
C1
6. As m we get (2).

Examples.
3.5.1
Homework
Exercises: 3.5[2,4,6,8,10]
3.6. PROPERLY DIVERGENT SEQUENCES
3.6
49
Properly Divergent Sequences

Definition 3.6.1
Examples.
Theorem 3.6.1
Proof:
1.

Theorem 3.6.2
Proof:
1.

Theorem 3.6.3
Proof:
1.
Examples.
50
3.7
Introduction to Series
Definition 3.7.1
Examples.
Theorem 3.7.1 The n-th Term Test.

Proof:
1.
Theorem 3.7.2 Cauchy Criterion for Series.
Theorem 3.7.3 The n-th Term Test.

Proof:
1.
Examples.
3.7.1
Comparison Tests
Theorem 3.7.4 Comparison Test.
Proof:
1.
Theorem 3.7.5 Limit Comparison Test.

Proof:
1.
Examples.
Chapter 4
Limits
4.1
Limits of Functions
Definition 4.1.1
Let A R. A point c R is a cluster point (or
accumulation point, or limit point) of A if for every > 0 there exists
a point x A such that x , c and |x c| < .
A point c R is a cluster point of A if every -neighborhood B (c) of c
contains an element of A distinct from c.
The point c does not have to be an element of A.

Examples.
Let X = (xn ) be a sequence and c R. We say
X , c if xn , c for any n N.
Theorem 4.1.1 Let A R. A number c R is a cluster point of A if

and only if there exists a sequence X in A such that X , c and lim X = c.
Proof: Easy.
Examples.
51
CHAPTER 4. LIMITS
52
4.1.1
The Definition of the Limit

Definition 4.1.2 Let A R, c be a cluster point of A, f : A R be a
function, and L R be a real number. We say that L is a limit of f at c
if for any > 0 there exists > 0 such that
if |x c| < and x , c then | f (x) L| < .

Then we say that f converges to L at c.
If the limit of f does not exists then we say that f diverges at c.
Notation.
lim f (x) = L,
xc
lim f = L,
xc
xc
f (x) L
Theorem 4.1.2 Let A R, f : A R and c be a cluster point of A.

Then f can have only one limit at c.
Proof: Easy.
Theorem 4.1.3 Let A R, c be a cluster point of A, f : A R be a

function, and L R be a real number. Then lim xc f (x) = L if and only
if for any -neighborhood B (L) of L, there exists a -neighborhood
B (c) of c such that for any x A, x , c, if x B (c) then f (x) B (L).
Proof: Exercise.
Examples.
4.1.2
Sequential Criterion for Limits

Theorem 4.1.4 Sequential Criterion. Let A R, c be a cluster point
of A and f : A R. Then lim xc f (x) = L if and only if for every
sequence X in A such that X , c and lim X = c, the sequence f (X)
converges to L, i.e. lim f (X) = L.
Proof:
1. (I). Suppose lim xc f (x) = L.
4.1. LIMITS OF FUNCTIONS
53
2. Then, > 0, > 0 such that x A, x , c, if |x c| < then

| f (x) L| < .
3. Let X = (xn ) be a sequence in A such that X , c and lim X = c.
4. Then, for any > 0, K N such that n K, |xn c| < .
5. Therefore, | f (xn ) L| < .
6. Thus lim f (X) = L.
7. (II). Suppose that for every sequence X in A such that X , c and
lim X = c the sequence f (X) converges to L.
8. Suppose that f does not converge to L at c.
9. Then there exists 0 such that > 0 x A such that x , c and
|x c| < and | f (x) L| 0 .
10. Then n N, xn A, xn , c such that |xn c| <
1
n
but | f (xn ) L| 0 .
11. Therefore f (X) does not converge to L (contradiction).

4.1.3
Divergence Criteria
Theorem 4.1.5 Divergence Criteria. Let A R, c be a cluster point
of A, L R, and f : A R.
1. The function f does not have limit L at c if and only if there exists
a sequence X in A such that X , c and lim X = c but the sequence
f (X) does not converge to L.
2. The function f diverges at c if and only if there exists a sequence
X in A such that X , c and lim X = c but the sequence f (X)
diverges.
Proof: Exercise.
Examples.
4.1.4
Homework
Exercises: 4.1[3,6,11(a),12,14]
CHAPTER 4. LIMITS
54
4.2
Limits Theorems
Definition 4.2.1 Let A R, c be a cluster point of A, f : A R. The
function f is said to be bounded on a neighborhood of c if there exists
a -neighborhood B (c) of c and a constant M > 0 such that
if x A B (c), then | f (x)| M.

That is if there exist > 0 and M > 0 such that
if x A and |x c| < , then | f (x)| M.
Theorem 4.2.1 Let A R, c be a cluster point of A and f : A R. If

f has a limit at c, then f is bounded on a neighborhood of c.
Proof:
1. Let L = lim xc f . Let = 1.
2. We have
| f (x)| L | f (x) L| < 1
3. If c A, let M = max{| f (c)|, L + 1}.
4. Then x A,
| f (x)| M.
Examples.
Definition 4.2.2 Let A R, a R, and f, g, h, : A R be functions
such that h(x) , 0 and (x) 0 for all x A. The sum f + g, the
difference f g, thepproduct f g, the multiple a f , the absolute value
| f |, the square root f and the quotient f /h are the functions defined
by: x A
( f g)(x) = f (x) g(x),

(a f )(x) = a f (x),
| f |(x) = | f (x)|,
( f g)(x) = f (x)g(x),
!
f (x)
f
(x) =
,
h
h(x)
p
(x) = (x).
4.2. LIMITS THEOREMS
55
Theorem 4.2.2 Let A R, c be a cluster point of A, a R, and f, g, h :

A R be functions such that h(x) , 0 for all x A and lim xc f (x) =
L, lim xc g(x) = M, lim xc = K, lim xc h(x) = H , 0. Then
lim( f g) = L M
lim( f g) = LM,
xc
xc
lim(a f ) = aL,
xc
lim | f | = |L|,
xc
!
f
L
lim
= .
xc h
H
lim = K
xc
Proof: Easy. Use sequential criterion.

Remarks.
If lim xc h = 0, the theorem is not valid.
Similarly, we have
lim( f (x))n = Ln
xc
c does not have to be in the domain.

If f is unbounded at c, then it does not have a limit at c.
Examples.
Theorem 4.2.3 Let A R, a, b R, c be a cluster point of A, and
f : A R be a function such that x A, x , c
a f (x) b.
Then if f has a limit at c, then

a lim f b.
xc
Proof:
1. Use a sequence X in A such that X , c and lim X = c

CHAPTER 4. LIMITS
56
Theorem 4.2.4 Squeeze Theorem. Let A R, c be a cluster point of

A, and f, g, h : A R be functions such that x A, x , c
f (x) g(x) h(x).
Then if lim xc f = lim xc h, then

lim g = lim f = lim h.
xc
xc
xc
Proof: Exercise. By sequences.

Examples.
Theorem 4.2.5 Let A R, c be a cluster point of A, and f : A R be
a function such that
lim f > 0.
xc
Then there exists > 0 such that for any x A, x , c, if |x c| < , then
f (x) > 0.
That is, there is a -neighborhood B (c) of c such that f (x) > 0 for all
x A B (c), x , c.
Proof: Easy.
Note. f (c) may be negative.

Extension of the Limit Concept.
Right-hand limit.
Left-hand limit.
Limit at infinity.
4.2.1
Homework
Exercises: 4.2[4,5,11(b,c),12]
Chapter 5
5.1
Definition 5.1.1
Let A R, c A, and f : A R. Then f is
continuous at c if for any > 0 there exists > 0 such that x A,
if |x c| < then | f (x) f (c)| < .

If f is not continuous at c, then f is said to be discontinuous at c.
Theorem 5.1.1 Let A R, c A, and f : A R. Then f is continuous
at c if and only if for any -neighborhood B ( f (c)) of f (c) there exists a
-neighborhood B (c) such that
if x A B (c) then f (x) B ( f (c)),

that is
f (A B (c)) B ( f (c))
Remarks.
If c A is not a cluster point of A, then f is automatically continuous at c.
If c is a cluster point of A then f is continuous at c if and only if
f (c) = lim f (x)
xc
57
CHAPTER 5. CONTINUOUS FUNCTIONS
58
Theorem 5.1.2 Sequential Criterion for Continuity. Let A R, c

A, and f : A R. Then f is continuous at c if and only if for any
sequence X in A converging to c, the sequence f (X) converges to f (c).
Theorem 5.1.3 Discontinuity Criterion. Let A R, c A, and

f : A R. Then f is discontinuous at c if and only if there exists
a sequence X in A such that X converges to c but the sequence f (X)
does not converge to f (c).
Definition 5.1.2
Let A R, B A, and f : A R. Then f is
continuous on the set B if f is continuous at every point of B.
Examples.
The Dirichlet Function f : R R defined by
(
0, if x is irrational
f (x) =
1, if x is rational
is discontinuous at every point.
Proof: Use sequences.

The Thomae Function f : R R defined by
0, if x is irrational
p
1
, if x = is rational, q > 0
f (x) =
q
q
and
p
and
q are relatively prime
1, if x = 0
is discontinuous at every rational number and continuous at every irrational
number.
Proof:
1. Use sequences for rational numbers.
2. For any irrational number c, given > 0 there exists a -neighborhood
B (c) of c such that it does not contain any rational numbers x = p/q
with q < 1/.
5.1. CONTINUOUS FUNCTIONS
59
3. Therefore, x B (c), | f (x)| < .

4. So, lim xc f = f (c) = 0.
5. Thus f is continuous at every irrational number.

Continuous extension of a function. Let f : A R and c be a cluster
point of A such that c < A. If f has a limit at c, then one can define a
continuous extension F : A {c} R R of f by
(
f (x),
if x A
F(x) =
lim xc f, if x = c
5.1.1
Homework
Exercises: 5.1[3,7,11,12,13]
60
5.2
Combinations of Continuous Functions
Let A R and f : A R. We say that

f , 0 if x A, f (x) , 0.
We say that
f 0 if x A, f (x) 0.
Let A R and f : A R. The absolute value of f is a function | f | : A R
defined by
| f |(x) = | f (x)|,
x A.
p
Let A R, f : A R and f 0. The square root of f is a function
f : A R defined by
p
p
f (x) = f (x),
x A.
Theorem 5.2.1 Let A R, a R, c A, and f, g, h, : A R be

continuous functions at c such that h , 0 and 0. Then the functions
f g, f g, a f , f /h, | f | and are continuous at c.

Proof:
1. If c is an isolated point of A, then it is automatic.
2. If c is a cluster point of A, then the assertions follow from the limit
theorems.
Theorem 5.2.2 Let A R, a R, and f, g, h, : A R be continuous

functions on A such that h , 0 and 0. Then the functions f g, f g,
a f , f /h, | f | and are continuous on A.
Definition of Quotients. Let A R and f, h : A R. Let

A0 = {x A | h(x) , 0}.
The quotient f /h : A0 R is defined for any x A0 by
!
f
f (x)
.
(x) =
h
h(x)
Then if f and h are continuous on A0 , then f /h is continuous on A0 .
5.2. COMBINATIONS OF CONTINUOUS FUNCTIONS
61
Examples.
Polynomials.
Rational Functions.
Trigonometric Functions.
5.2.1
Composition of Continuous Functions

Theorem 5.2.3 Let A, B R, f : A B and g : B R. Let c A and
b = f (c) B. Then if f is continuous at c and g is continuous at b, then
the composition h = g f : A R is continuous at c.
Proof:
1. Let c A, b = f (c) and a = g(b).
2. Let > 0.
3. There exists > 0 such that if y B B (b), then g(y) B (a).
4. There exists > 0 such that if x A B (c), then f (x) B (a).
5. Therefore, if x A B (c), then h(x) = (g f )(x) = g( f (x)) B (a).
6. Thus, h = g f is continuous at c A.
Theorem 5.2.4 Let A, B R, f : A B and g : B R. Then if

f is continuous on A and g is continuous on B, then the composition
h = g f : A R is continuous on A.
Proof: Follows from the previous theorem.
Examples.
5.2.2
Homework
Exercises: 5.2[5,6,10,12,13]
62
5.3
Continuous Functions on Intervals

Definition 5.3.1 Let A R and f : A R. We say that f is bounded
on A if there exists M R such that | f | M, that is x A, | f (x)| M.
We say that f is unbounded on A if M > 0 there exists x A such

that | f (x)| > M.
A function is bounded on a set if its range is bounded and unbounded if its
range is unbounded.
Theorem 5.3.1 Boundedness Theorem. Let I = [a, b] be a closed

bounded interval and f : I R be continuous on I. Then f is bounded
on I.
Proof:
1. By contradiction. Suppose f is not bounded.
2. Then there exists a sequence X = (xn ) in I such that
f (xn ) > n
3. Since X is bounded, there exists a convergent subsequence X 0 = (xnk ).
4. Since I is closed c = lim X 0 I.
5. Since f is continuous lim f (X 0 ) = f (c).
6. Hence f (X 0 ) is bounded, which is a contradiction since
| f (xnk )| > nk k.
Examples.
5.3. CONTINUOUS FUNCTIONS ON INTERVALS
5.3.1
63
The Maximum-Minimum Theorem

Definition 5.3.2
Let A R and f : A R. We say that f has
an absolute maximum on A if there exists x1 A, called an absolute
maximum point of f on A, such that
f (x1 ) f (x) for any x A.
We say that f has an absolute minimum on A if there exists x2 A,

called an absolute minimum point of f on A, such that
f (x2 ) f (x) for any x A.
Remarks.
x1 and x2 must be in A.
x1 and x2 are not unique, in general.
Example.
Theorem 5.3.2 Maximum-Minimum Theorem. Let I = [a, b] be a

closed bounded interval and f : I R be continuous on I. Then f has
an absolute maximum and an absolute minimum on I.
Proof:
1. Let u = sup f (I).
2. Claim: c I such that u = f (c).
3. There exists a sequence X = (xn ) in I such that n N
u
1
< f (xn ) u
n
4. There exists a convergent subsequence X 0 = (xnk ).

5. Since I is closed c = lim X 0 I.
6. Since f is continuous, lim f (X 0 ) = f (c).
7. On the other hand lim f (X 0 ) = u.
8. Therefore, f (c) = sup f (I) and c is a maximum point of f on I.
9. Similarly, for the minimum point.

64
5.3.2
Bisection Method
Theorem 5.3.3 Location of Roots Theorem. Let I = [a, b] and f :
I R be continuous on I. If f (a) < 0 < f (b) or f (b) < 0 < f (a), then
there exists c (a, b) such that f (c) = 0.
Proof:
1. Let f (a) < 0 < f (b).
2. Define a nested sequence (Ik = [ak , bk ]) of intervals defined by bisection so that
Ik+1 Ik ,
bk ak =
ba
,
2k1
f (ak ) < 0 < f (bk )
3. If at some step we find a midpoint pk = (ak +bk )/2 such that f (pn ) = 0,
then just let c = pn . Otherwise, we get an infinite sequence of nested
intervals.
4. Then there exists c I such that
lim an = lim bn = c
5. Since f is continuous,
f (c) = lim f (an ) 0 and f (c) = lim f (bn ) 0
6. Thus f (c) = 0.

Example.
5.3.3
Bolzanos Theorem
Theorem 5.3.4 Bolzanos Intermediate Value Theorem. Let I be an
interval and f : I R be continuous on I. Let a, b I and there exist
k R such that f (a) < k < f (b). Then there exists c I between a and
b such that f (c) = k.
Proof:
1. Let a < b.
5.3. CONTINUOUS FUNCTIONS ON INTERVALS
65
2. Let g(x) = f (x) k.

3. Use Location of Roots Theorem.

Corollary 5.3.1 Let I = [a, b] be a closed bounded interval and f :
I R be continuous on I. If there exists a k R such that
inf f (I) k sup f (I),

then there exists c I such that f (c) = k.
Proof:
1. Follows from the Maximum-Minimum Theorem and Bolzanos Theorem.
Theorem 5.3.5 Let I be a closed bounded interval and f : I R be

continuous on I. Then the range f (I) of f is a closed bounded interval.
A continuous image of a closed and bounded interval is a closed and
bounded interval.
Proof:
1. Let m = inf f (I) and M = sup f (I).
2. Then m, M f (I) and f (I) [m, M].
3. Let k [m, M].
4. Then c I such that f (c) = k.
5. So k f (I) and [m, M] f (I).
6. Thus f (I) = [m, M].
Remarks.
Closed and bounded intervals are compact sets.
Continuous functions preserve compactness.
Not true for not closed or unbounded intervals.
66
Examples.
Theorem 5.3.6 Preservation of Intervals Theorem. Let I be an interval and f : I R be continuous on I. Then the range f (I) of f is an
interval.
A continuous image of an interval is an interval.
Proof:
1. Let u, v f (I) and u < v.
2. Then a, b I such that f (a) = u and f (b) = v.
3. Then k (u, v) c I such that f (c) = k.
4. So, [u, v] f (I).
5. Thus, f (I) is an interval.
5.3.4
Homework
Exercises: 5.3[1,3,6,13,15]
5.4. UNIFORM CONTINUITY
5.4
67
Uniform Continuity
Definition 5.4.1 Let A R and f : A R. Then f is uniformly
continuous on A if > 0 > 0 such that x, y A
if |x y| < , then | f (x) f (y)| < .
Remarks.
= () depends on but not on x, y!
If f is just continuous, but not uniformly continuous, then = (, x) depends on both and x A.
Let () = inf xA (, x). If () > 0, then f is uniformly continuous. If
() = 0, then f is continuous but not uniformly continuous.
Not every continuous function is uniformly continuous.
Examples.
Theorem 5.4.1 Nonuniform Continuity Criteria. Let A R and f :
A R. The following are equivalent statements:
1. f is not uniformly continuous on A.
2. there exists 0 > 0 such that > 0 x, y A such that

|x y| < and | f (x) f (y)| 0 .
3. there exists 0 > 0 and two sequences X and Y in A such that
lim(X Y) = 0 and | f (X) f (Y)| 0 .
Proof: Exercise.
Examples.
Theorem 5.4.2 Uniform Continuity Theorem. Let I be a closed

bounded interval and f : I R be continuous on I. Then f is uniformly continuous on I.
Proof:
68
1. By contradiction. Suppose f is not uniformly continuous.

2. Then 0 > 0 and two sequences X = (xn ) and Y = (yn ) in I such that
n N,
|xn yn | <
1
n
but | f (xn ) f (yn )| 0 .
3. Since I is bounded, there is a convergent subsequence X 0 = (xnk ).

4. Since I is closed, z = lim X 0 I.
5. Since
|ynk z| |ynk xnk | + |xnk z|
as k the subsequence Y 0 = (ynk ) also converges to z.
6. Since f is continuous
f (z) = lim f (X 0 ) = lim f (Y 0 ).
7. This contradicts | f (xn ) f (yn )| 0 .
5.4.1
Lipshitz Functions
Definition 5.4.2 Let A R and f : A R. Then f is a Lipshitz
function if K > 0 such that x, y A
| f (x) f (y)| K|x y|.
f is Lipshitz if and only if the slope of any secant line of the graph of f is
bounded.
For a differentiable function, Lipshitz means that the derivative is bounded.
More generally, Lipshitz condition of order > 0
| f (x) f (y)| K|x y|
Theorem 5.4.3 Let A R and f : A R be Lipshitz. Then f is

uniformly continuous.
Proof:
5.4. UNIFORM CONTINUITY
69
1. Easy.

Not all uniformly continuous functions are Lipshitz.
Examples.
5.4.2
The Continuous Extension Theorem

Theorem 5.4.4 Let A R and f : A R be uniformly continuous.
Then if X is a Cauchy sequence in A, then f (X) is a Cauchy sequence.
Proof:
1. Let > 0.
2. Let X = (xn ) be a Cauchy sequence in A.
3. There is > 0 such that x, y A,
if x y| < , then | f (x) f (y)| < .
4. There is K N such that n, m K,
|xn xm | < .
5. Thus, n, m K,
| f (xn ) f (xm )| <
and f (X) is Cauchy.
Theorem 5.4.5 Continuous Extension Theorem. A function f :

(a, b) R on an open interval (a, b) is uniformly continuous if and
only if it can be extended to a continuous function F : [a, b] R on the
closed interval [a, b].
Proof:
1. (I). Obviously, if F is continuous then its restriction f is uniformly
continuous.
2. (II). Suppose f is uniformly continuous.
3. Claim: The limit lim xa f (x) exists.
70
4. Let X = (xn ) be a sequence in (a, b) such that lim X = a.

5. Then X is Cauchy and f (X) is Cauchy.
6. So, f (X) converges.
7. Let Y = (yn ) be any other sequence in (a, b) such that lim Y = a.
8. Then lim(X Y) = 0.
9. Since f is uniformly continuous,
lim f (Y) = lim[ f (Y) f (X)] + lim f (X) = lim f (X).
10. Therefore, lim xa f (x) exists.
11. Similarly, lim xb f (x) exists.
12. Define F : [a, b] R by
lim xa f (x), x = a
f (x),
x (a, b)
F(x) =
lim
xb f (x), x = b
13. Then F is continuous on [a, b].

Examples.
5.4.3
Homework
Exercises: 5.4[2,3,6,7,8,12]
5.5. CONTINUITY AND GAUGES
5.5
71
Continuity and Gauges

Definition 5.5.1
Let I = [a, b] be an interval. Let xk I, (k =
0, . . . , n), be a set of points in I, called the partition points, such that
a = x0 < < xk1 < xk < . . . xn = b
and Ik = [xk1 , xk ], (k = 1, . . . , n), be a set of non-overlapping closed
subintervals of I whose union is [a, b].
The set P = {Ik | (k = 1, . . . , n)} of the intervals Ik is a partition of the

interval I.
Let tk Ik , (k = 1, . . . , n), be fixed points in each interval Ik , called the
tags. Then the set of ordered pairs
P = {(Ik , tk ) | k = (1, . . . , n)}
is a tagged partition of I.
Definition 5.5.2 Let I be an interval. A gauge on I is a strictly positive function : I R+ on I.
Let P = {(Ik , tk )|k = (1, . . . , n)} be a tagged partition of I. Then P is

-fine if
Ik [tk (tk ), tk + (tk )],
5.5.1
(k = 1, . . . , n) .
Lemma 5.5.1 Let I = [a, b] be an interval, x I, and : I R+ be

a gauge on I. The every tagged -fine partition P of I has a tag tk such
that |x tk | (tk ).
Existence of -fine Partitions

Theorem 5.5.1 Let I = [a, b] and : I R+ be a gauge on I. Then
there exists a -fine partition of I.
72
5.6
Monotone and Inverse Functions

Definition 5.6.1
Let A R and f : A R. We say that:
f is increasing on A if x, y A,
if x < y, then f (x) f (y).
f is strictly increasing on A if x, y A,
if x < y, then f (x) < f (y).
f is decreasing on A if x, y A,
if x < y, then f (x) f (y).
f is strictly decreasing on A if x, y A,
if x < y, then f (x) > f (y).
f is monotone on A if it is either increasing or decreasing on A.
f is strictly monotone on A if it is either strictly increasing or strictly
decreasing on A.
If f is increasing, then ( f ) is decreasing and vice versa.

Recall the left- and the right-handed limits.
Theorem 5.6.1 Let I be an interval and f : I R be increasing on I.
Let c I be an interior point of I. Then:
1. lim xc f = sup f (I (, c)),

2. lim xc+ f = inf f (I (c, )).
Proof:
1. (1). Let x I and x < c.
2. The set f (I (, c)) is bounded by f (c).
3. Let L = sup f (I (, c)).
4. Let > 0.
5.6. MONOTONE AND INVERSE FUNCTIONS
73
5. Then y0 I, y0 < c, such that

L < f (y0 ) L.
6. Let = c y0 .
7. Then for any y between y0 and c
L < f (y0 ) f (y) L.
8. So, 0 < c y < implies | f (y) L| < .
9. Thus, lim xc f = L.
10. (2). Similarly.

Corollary 5.6.1 Let I be an interval and f : I R be increasing on
I. Let c I be an interior point of I. Then the following are equivalent
statements:
1. f is continuous at c.
2. f (c) = lim xc f = lim xc+ f .
3. f (c) = sup f (I (, c)) = inf f (I (c, )).
Proof: Exercise.
Remarks.
Let I = [a, b] and f : I R be monotone on I. Then
f is continuous at a if and only if f (a) = lim xa+ f .
Similarly,
f is continuous at b if and only if f (b) = lim xb f .
74
Definition 5.6.2 Let I be an interval and c I be an interior point of I.

Let f : I R. Then the jump of f at c is
j f (c) = lim+ f lim f
xc
xc
The jump of f at a is
j f (a) = lim+ f f (a)
xa
The jump of f at b is
j f (b) = f (b) lim f
xb
For an increasing function the jump is equal to

j f (c) = inf f (I (c, )) sup f (I (, c))
Remark. The jump of f at c does not depend on the value f (c).
Theorem 5.6.2 Let I be an interval and f : I R be increasing on I.

Let c I. Then f is continuous at c if and only if j f (c) = 0.
Proof:
1. Follows directly from the definitions.
Theorem 5.6.3 Let I be an interval and f : I R be monotone on I.

Let D I be the set of points at which f is discontinuous. Then D is a
countable set.
Proof:
1. Let f be increasing and I = [a, b].
2. Then c I, j f (c) 0.
3. Let
D = {x I | j f (x) > 0}
and
)
1
Dn = x I | j f (x)
,
n
nN
5.6. MONOTONE AND INVERSE FUNCTIONS

4. Then
D=
75
Dn
n=1
5. Claim: the set Dn is finite.

6. Suppose Dn is infinite.
7. Let P = {x1 , . . . xm } Dn be a set of points in Dn such that
a x1 < < xm b
8. We have
m X
j f (xk ) f (b) f (a).
n
k=1
m
9. By choosing m > n( f (b) f (a)) we get a contradiction.

10. Thus Dn is finite, and the set D is countable.
5.6.1
Inverse Functions
Theorem 5.6.4 Continuous Inverse Theorem. Let I, J R be intervals and f : I J be strictly monotone and continuous function
from I onto J = f (I). Then f has an inverse and the inverse function
f 1 : J I is strictly monotone and continuous function from J onto I.
Proof:
1. Let f be strictly increasing.
2. Then f is an bijection and f 1 exists.
3. Claim: f 1 is strictly increasing.
4. Let y1 , y2 J and y1 < y2 .
5. Then x1 , x2 I such that f (x1 ) = y1 and f (x2 ) = y2 .
6. Since f is strictly increasing, x1 < x2 .
7. Therefore, f 1 (y1 ) = x1 < x2 = f 1 (y2 ).
8. Claim: f 1 is continuous on J.
9. Suppose that f 1 is discontinuous at some c J.
76
10. Then limyc f 1 < limyc+ f 1 .

11. Then for any x I such that x , f 1 (c) and limyc f 1 < x <
limyc+ f 1 there is no y J such that f 1 (y) = x.
12. Therefore, x < I, which is a contradiction.
5.6.2
Homework
Exercises: 5.6[2,5,7,9,10,12]
Chapter 6
Differentiation
6.1
The Derivative
Definition 6.1.1
Let I be an interval, f : I R and c I. Let
: I R be a function defined by
f (x) f (c)
, x,c
xc
,
(x) =
(c),
x=c
where (c) is an arbitrary real number. The function f is differentiable

at c if the function has a limit at c. In this case the limit lim xc is
denoted by f 0 (c) and called the derivative of f at c.
That is, the function f is differentiable at c if there exists a real number
f 0 (c) such that > 0 > 0 such that x I, x , c,

f (x) f (c)

0
if |x c| < , then
f (c) < .
xc
Notation. The derivative is also denoted by

df
,
dx
or just
D f.
Remarks.
It is allowed that c is the endpoint of the interval.
77
CHAPTER 6. DIFFERENTIATION
78
One can define the derivative of functions f : A R on any domain A with

a cluster point c.
Theorem 6.1.1 Caratheodorys Theorem. Let I be an interval, c I,

f : I R. Let : I R be a function defined by
f (x) f (c)
, x,c
xc
(x) =
(c),
x=c
Then f is differentiable at c if and only if the value (c) can be chosen
such that is continuous at c, that is (c) = lim xc (x). In this case
f 0 (c) = (c).
Proof:
1. (I). If f is differentiable, choose (c) = f 0 (c). Then is continuous.
2. (II). If is continuous at c, then (c) = lim xc (x) = f 0 (c).
Let I1 I be the set of points where the function f is differentiable. The

derivative f 0 : I1 R is a function on I1 .
Theorem 6.1.2 Let I be an interval, f : I R, and c I. If f is

differentiable at c, then it is continuous at c.
Proof: Easy.
Not all continuous functions are differentiable.

Example.
There are functions that are continuous everywhere but are not differentiable
at any point.
Example. (Weierstrass function).
6.1. THE DERIVATIVE
79
Theorem 6.1.3 Let I be an interval, c I, a R, and f, g, h : I R

be differentiable at c, and h(c) , 0. Then
1. The function a f is differentiable at c and
(a f )0 (c) = a f 0 (c).
2. The function f + g is differentiable at c and
( f + g)0 (c) = f 0 (c) + g0 (c).
3. The function ( f g) is differentiable at c and

( f g)0 (c) = f 0 (c)g(c) + f (c)g0 (c).
4. The function ( f /h) is differentiable at c and
!0
f
f 0 (c)h(c) f (c)h0 (c)
(c) =
.
h
(h(c))2
Proof: Formal computation of the limits.

Corollary 6.1.1 Let I be an interval, c I, and f, f1 , . . . , fn : I R be

differentiable at c. Then
( f1 fn )0 (c) = f10 (c) f2 (c) fn (c) + + f1 (c) fn1 (c) fn0 (c).
In particular,
( f n )0 (c) = n f n1 (c) f 0 (c).
Examples.
80
6.1.1
The Chain Rule

Theorem 6.1.4 Chain Rule. Let I and J be intervals and f : I R
and g : J R be functions such that f (I) J. Let h = g f : I R be
the composite function. Let c I and d = f (c) J, f be differentiable
at c and g be differentiable at d. Then h is differentiable at c and
h0 (c) = g0 (d) f 0 (c).
Proof:
1. Let : I R be the function defined by
f (x) f (c)
, x,c
xc
(x) =
f 0 (c),
x=c
2. Similarly, let : J R be the function defined by
g(y) g(d)
, y , f (c)
xc
(y) =
g0 (d),
y=d
3. Then the function h = ( f ) : I R defined by
h(x) = ( f (x))(x)
is continuous at c and
h(c) = g0 (d) f 0 (c).
Examples.
f is not differentiable but continuous.
f is differentiable everywhere but f 0 is not continuous.
f is differentiable, f 0 is continuous but f 0 is not differentiable.
6.1. THE DERIVATIVE
6.1.2
81
Inverse Functions
Theorem 6.1.5 Let I be an interval, f : I J be strictly monotone
and continuous function from I onto J. Let f 1 : J I be the strictly
monotone and continuous function inverse to f . Let c I and d =
f (c) J. If f is differentiable at c, then f 1 is differentiable at d and
( f 1 )0 (d) =
1
.
f 0 (c)
Proof:
1. Let : I R be the function defined by
f (x) f (c)
, x,c
xc
(x) =
f 0 (c),
x=c
2. Then (c) = f 0 (c) , 0.
3. Thus, there exists a -neighborhood B (c) = (c , c + ) of c such that
x B(c) I,
(x) , 0.
4. Let U = f (B(c) I).
5. Let g = f 1 . Then y U, f (g(y)) = y. In particular g(d) = c.
6. Let h = 1/( g) : U R.
7. Then y U
g(y) g(d)
, y,d
yd
1
h(y) =
=
1
(g(y))
,
y=d
(c)
8. It is continuous at d and
g0 (d) =
1
1
= 0 .
(c)
f (c)
Examples.
82
6.1.3
Homework
Exercises: 6.1[4,9,10,11(c),13]
6.2. THE MEAN VALUE THEOREM
6.2
83
The Mean Value Theorem

Definition 6.2.1 Let I be an interval, c I, and f : I R. Then f has
a relative maximum (minimum) at c if there exists a -neghborhood
B (c) of c such that x B (c) I,
f (x) f (c),
( f (x) f (c)).
We say that f has a relative extremum at c if it has either a relative

maximum or relative minimum at c.
Theorem 6.2.1 Interior Extremum Theorem. Let I be an interval,

c I be an interior point of I, and f : I R. Let f be differentiable at
c and have a relative extremum at c. Then f 0 (c) = 0.
Proof:
1. We have
f 0 (c) = lim
xc
f (x) f (c)
xc
2. Show that it is impossible to have f 0 (c) > 0 and f 0 (c) < 0.

Corollary 6.2.1 Let I be an interval, c I be an interior point of I, and

f : I R. Let f be continuous on I and have a relative extremum at c.
Then either f is not differentiable at c or f 0 (c) = 0.
Theorem 6.2.2 Rolles Theorem. Let I = [a, b] be a closed and
bounded interval, and f : I R. Let f be continuous on the
closed interval [a, b] and differentiable on the open interval (a, b), and
f (a) = f (b) = 0. Then there exists a point c (a, b) such that
f 0 (c) = 0.
Proof:
1. Let
M = sup f (I),
m = inf f (I) .
2. Then
m 0 M.
84
3. If both m = M = 0 then f = 0.
4. If M > 0, then there is c (a, b) such that f (c) = M (maximum).

5. If m < 0, then there is d (a, b) such that f (d) = m (minimum).
6. Thus, there is a point in (a, b) where f 0 vanishes.
Theorem 6.2.3 Mean Value Theorem. Let I = [a, b] be a closed and

bounded interval, and f : I R. Let f be continuous on the closed
interval [a, b] and differentiable on the open interval (a, b). Then there
exists a point c (a, b) such that
f 0 (c) =
f (b) f (a)
.
(b a)
Proof:
1. Consider the function
g(x) = f (x) f (a)
f (b) f (a)
(x a)
(b a)
and apply Rolles Theorem.

2. Then g0 (c) = 0 and the theorem follows.
Theorem 6.2.4 Let I = [a, b] be a closed and bounded interval, and f :

I R. Let f be continuous on the closed interval [a, b], differentiable
on the open interval (a, b), and x (a, b)
f 0 (x) = 0.
Then f is constant on I.
Proof:
1. Let x (a, b).
2. Then there is c (a, x) such that
f 0 (c) =
f (x) f (a)
=0
xa
85
3. Thus, f is constant.
Corollary 6.2.2 Let I = [a, b] be a closed and bounded interval, and

f, g : I R. Let f, g be continuous on the closed interval [a, b],
differentiable on the open interval (a, b), and x (a, b)
f 0 (x) = g0 (x).
Then there exists a constant C such that f = g + C on I.
Proof: Easy.

Theorem 6.2.5 Let I be an interval, and f : I R. Let f be differentiable on I. Then
1. f is increasing on I if and only if f 0 is nonnegative on I.

2. f is decreasing on I if and only if f 0 is nonpositive on I.
Proof:
1. If f is increasing, then f 0 0.
2. Now, suppose that f 0 0.
3. Let x1 , x2 I and x1 < x2 .
4. Then there is c (x1 , x2 ) such that
f 0 (c) =
f (x2 ) f (x1 )
0.
x2 x1
5. Thus, f is increasing.
Definition 6.2.2 Let I be an interval, c I, and f : I R. Then f is

increasing at c if there exists a -neghborhood B (c) of c such that f is
increasing on B (c) I.
Remarks.
Even if f is strictly increasing on I then f 0 is still nonnegative on I (there
may be points where f 0 (c) = 0).
86
However, if f 0 is strictly positive on I, then f is strictly increasing on I.

Positivity of f 0 (c) > 0 at a point c is not enough for f to be increasing at c.
That is f 0 (c) > 0 does not mean that f increases at c.
One needs positivity of f 0 on a neighborhood of a point.
Example.
, x,0
x + 2x sin
x
f (x) =
0,
x=0
Then f 0 (0) = 1 > 0 however f changes sign in every neighborhood of 0 and

f is not increasing at 0.
Theorem 6.2.6 First Derivative Test for Extrema. Let I = [a, b] be
an interval, c be an interior point of I and f : I R. Let f be continuous on I and differentiable on (a, c) and (c, b). Then:
1. If there exists a > 0 such that (c , c + ) I and
f 0 (x) 0 for x (c , c) and f 0 (x) 0 for x (c, c + ),
then f has a relative maximum at c.

2. If there exists a > 0 such that (c , c + ) I and
f 0 (x) 0 for x (c , c) and f 0 (x) 0 for x (c, c + ),
then f has a relative minimum at c.
Proof:
1. Let x (c , c).
2. Then there is d (x, c) such that
f 0 (d) =
f (c) f (x)
0
xc
3. So, f (x) f (c).

4. Similarly, f (c) f (y) for any c < y < c + .
87
5. Thus, f has a maximum at c.

The converse is not true!
Example.
4
4
, x,0
2x + x sin
x
f (x) =
0,
x=0
Then f has an absolute minimum at 0, f 0 (0) = 0, but f 0 changes sign in every neighborhood of 0, and, therefore, f is neither increasing nor decreasing
at 0.
6.2.1
Inequalities
Examples.
1.
ex 1 + x
2.
| sin x| |x|
3. Let x > 1 and > 1. Then
(1 + x) 1 + x
6.2.2
Intermediate Value Property of Derivatives

Lemma 6.2.1 Let I be an interval, c I, and f : I R. Let f be
differentiable at c. Then:
1. If f 0 (c) > 0, then there exists a > 0 such that
f (x) > f (c) for x (c, c + ) I.

2. If f 0 (c) < 0, then there exists a > 0 such that
f (x) > f (c) for x (c , c) I.
Proof:
88
1. Let f 0 (c) > 0.
2. We have
f 0 (c) = lim
xc
f (x) f (c)
>0
xc
3. Thus f (x) > f (c) for x > c.

Theorem 6.2.7 Darbouxs Theorem. Let I = [a, b] be an interval,

f : I R be differentiable on I. Let k be a number between f 0 (a) and
f 0 (b). Then there exists a point c (a, b) such that
f 0 (c) = k.
Proof:
1. Let f 0 (a) < k < f 0 (b).
2. Consider
g(x) = kx f (x)
3. Then g has maximum on I.
4. Since
g0 (a) = k f 0 (a) > 0
then the maximum is not at a.
5. Similarly, since
g0 (b) = k f 0 (b) < 0
then the maximum is not at b.
6. Thus maximum is at some c (a, b) where g0 (c) = 0.
Remarks.
The intermediate value property of the derivative is valid even if f 0 is not
continuous.
One can use Darboux theorem to check whether or not a given function can
be a derivative of another function.
Example.
6.2.3
89
Homework
Exercises: 6.2[7,10,12,13,17]
90
6.3
Taylors Theorem
Higher derivatives.
If f (n) exists, then f (k) exists for all k < n.
The spaces C n and C .
Definition 6.3.1 Let I be an interval and f : I R. Let x0 I, n N,

and f (k) exists for k = 1, . . . , n. The n-th Taylor polynomial for f at x0
is
n
X
f (k) (x0 )
Pn (x, x0 ) =
(x x0 )k .
k!
k=0
The function Rn defined by

Rn (x, x0 ) = f (x) Pn (x, x0 )
is the remainder.
Theorem 6.3.1 Taylors Theorem. Let I = [a, b], n N, x0 I,
f : I R, f (k) be continuous for k = 1, . . . , n on I, and f (n+1) exist on
(a, b). Then for any x I there exists c I between x and x0 such that
f (x) = Pn (x, x0 ) +
that is
Rn (x, x0 ) =
f (n+1) (c)
(x x0 )n+1 ,
(n + 1)!
f (n+1) (c)
(x x0 )n+1 .
(n + 1)!
Proof:
1. Let F : I R be defined by
F(t) = f (x) Pn (x, t) = Rn (x, t) .
2. Let G : I R be defined by
!n+1
xt
G(t) = F(t)
F(x0 )
x x0
(x t)n+1
Rn (x, x0 )
= Rn (x, t)
(x x0 )n+1
6.3. TAYLORS THEOREM
91
3. Then
G(x0 ) = G(x) = 0
4. Therefore, there exists c between x and x0 such that
(x c)n
0 = G0 (c) = F 0 (c) + (n + 1)
F(x0 )
(x x0 )n+1
5. We have
Rn (x, x0 ) = F(x0 )
6. Thus
and
f (n+1) (c)
F (c) =
(x c)n .
n!
0
f (n+1) (c)
Rn (x, x0 ) =
(x x0 )n+1 .
(n + 1)!
6.3.1
Relative Extrema
Theorem 6.3.2 Let I be an interval, f : I R, x0 I be an interior
point of I. Let n 2 and f (k) be continuous in a neighborhood of x0 for
k = 1, . . . , n and
f ( j) (x0 ) = 0
for
1 j n 1,
but
f (n) (x0 ) , 0 .
Then
1. If n is even and f (n) (x0 ) > 0, then f has a relative minimum at x0 .
2. If n is even and f (n) (x0 ) < 0, then f has a relative maximum at x0 .
3. If n is odd, then f has neither a relative minimum nor a relative
maximum at x0 .
Proof:
1. There is c between x and x0 such that
f (x) = f (x0 ) +
f (n) (c)
(x x0 )n
n!
2. Then f (n) (c) has the same sign as f (n) (x0 ).

3. The theorem follows.

92
6.3.2
Convex Functions
Definition 6.3.2 Let I be an interval and f : I R. Then f is convex
on I if t [0, 1] and x, y I there holds
f ((1 t)x + ty) (1 t) f (x) + t f (y) .
Example.
Remarks.
Convex function does not have to be differentiable.
A convex function on an open interval has the left and right derivatives at
any point.
A convex function on an open interval is necessarily continuous.
Theorem 6.3.3 Let I be an open interval and f : I R be a twice

differentiable function on I. Then f is convex if and only if
f 00 (x) 0
x I .
Proof:
1. (I). Suppose f is convex.
2. Let a I. We have

1
f
(a
+
h)
2
f
(a)
+
f
(a
h)
h0 h2
f 00 (a) = lim
3. Since f is convex
f (a)
1
1
f (a + h) + f (a h)
2
2
4. Thus
f 00 (a) 0 .
5. (II). Suppose f 00 (x) 0 for all x I.
6. Let x1 , x2 I and t [0, 1] be such that
x0 = (1 t)x1 + tx2 .
6.3. TAYLORS THEOREM
93
7. Then there exists c1 between x0 and x1 such that

f (x1 ) = f (x0 ) + f 0 (x0 )(x1 x0 ) +
1 00
f (c1 )(x1 x0 )2
2
and there exists c2 between x0 and x2 such that

f (x2 ) = f (x0 ) + f 0 (x0 )(x2 x0 ) +
1 00
f (c2 )(x2 x0 )2
2
8. Since f 00 0, we obtain
(1 t) f (x1 ) + t f (x2 ) f (x0 ) = f ((1 t)x1 + tx2 ) .
9. So, f is convex.
6.3.3
Homework
Exercises: 6.4[2,4,8,10,15]
94
Chapter 7
7.1
7.1.1
Partitions and Tagged Partitions

Definition 7.1.1 Let I = [a, b] be a closed and bounded interval. A
partition of I is a finite ordered set, an n-tuple,
P = (x0 , x1 , , xn1 , xn ),
of points in I such that
a = x0 < x1 < < xn1 < xn = b .
Let
Ik = [xk1 , xk ]
k = 1, . . . , n .
The norm (or mesh) of the partition P is the length of the largest interval Ik , that is
kPk = max |xk xk1 |
1kn
The tags {tk }nk=1 of the partition P are some fixed points in each subinterval Ik .
A tagged partition of I is the set of ordered pairs
P = {(Ik , tk )}nk=1 .
95
CHAPTER 7. THE RIEMANN INTEGRAL
96
Alternatively, the partition P can be also denoted by

P = {Ik }nk=1 .
Definition 7.1.2 Let I = [a, b], f : I R and P be a tagged partition
of I. The Riemann sum of f corresponding to P is
=
S ( f, P)
n
X
f (tk )(xk xk1 ) .
k=1
Examples.
7.1.2
Definition of Riemann Integral

Definition 7.1.3 Let I = [a, b] and f : I R. Then f is Riemann
integrable on I if there exists a L R such that for any > 0 there is a
> 0 such that for any tagged partition P of I,
< , then |S ( f, P)
L| < .
if kPk
The number L is the Riemann integral of f over I, denoted by

Rb
R
Rb
Rb
dx f .
f
(x)
dx
or
f
or
f
or
a
a
a
I
The set of all Riemann integrable functions on I is denoted by R(I).
Theorem 7.1.1 The Riemann integral of a Riemann integrable function

is uniquely defined.
Proof:
1. Assume L1 =
Rb
a
f and L2 =
Rb
a
2. Show that for any > 0

|L1 L2 | < .
3. So, L1 = L2 .

7.1. THE RIEMANN INTEGRAL
7.1.3
97
Examples
Step functions.
Functions with some isolated discontinuities.
Dirichlet function. (Not integrable).
Thomae function. (Integrable).
7.1.4
Properties of the Integral

Theorem 7.1.2 Let I = [a, b], f, g : I R, k R, and f, g R(I).
Then
1. k f R(I) and
kf = k
2. ( f + g) R(I) and
Z
( f + g) =
a
b
a
3. If f (x) g(x), x I, then

Z
f
a
g
a
f+
g.
a
Proof:
1. Easy.
2. Use

Z
Z !
Z
Z

f + g S ( f, P)
f + S ( f, P) g
S ( f + g, P)

3. Use
Z
S (g, P)
<
f < S ( f, P)
g+
98
7.1.5
Boundedness Theorem
Theorem 7.1.3 Let I = [a, b], f : I R and f R(I). Then f is
bounded on I.
Proof:
1. Suppose f is unbounded.
2. Choose the tag on the subinterval of the partition on which f is unbounded so that the Riemann sum becomes unbounded.
3. Let = 1.
if ||P||
< , then
4. Then there is such that P,
Z

S ( f, P)

f < = 1

or
|S ( f, P) < f + 1
5. Let k be the subinterval where f is unbounded.

6. We have
=
S ( f, P)
+ f (tk )(xk xk1 )
i,k
7. Thus

X
+ | f (tk )|(xk xk1 )
|S ( f, P)|

i,k
or
Z
X

| f (tk )|(xk xk1 ) + f + 1
i,k
8. We can choose tk to make the left side arbitrarily large, which leads to
a contradiction.

Example.
7.1.6
Homework
Exercises: 7.1[6(b),9,10,12,13]
7.2. RIEMANN INTEGRABLE FUNCTIONS
7.2
99
Riemann Integrable Functions

Theorem 7.2.1 Cauchy Criterion. Let I = [a, b] and f : I R. Then
f R(I) is integrable if and only if for any > 0 there exists > 0 such
< and kQk
< ,
that for any two tagged partitions P and Q of I, if kPk
then
S ( f, Q)|
< .
|S ( f, P)
Proof:
1. (I). Suppose f is integrable. Use triangle inequality.
Q,
if kPk
< and kQk
< , then |S ( f, P)
S ( f, Q)|
<
2. (II). Suppose P,
.
Q,
3. Define a decreasing sequence (n ) such that n N, n > 0 and P,

1
< and kQk
< , then |S ( f, P)
S ( f, Q)|
< .
if kPk
n
4. Define a sequence of partitions (Pn ) such that if kP n k < n .

n=1
5. Let an = S ( f, P n ).
6. If m > n, then kP m k < n and kP n k < n .
7. So, for m > n
|an am | <
1
.
n
8. Therefore, (an ) converges. Let A = lim an .

9. Then as m
|an A| <
10. Claim: A =
Rb
a
1
.
n
f.
11. Let > 0. Let K N.

if kQk
< K , then
12. Then Q,
A| |S ( f, Q)
aK | + |aK A| < 2 <
|S ( f, Q)
K
by choosing K > 2/.

Example.
100
Corollary 7.2.1 Criterion of Nonintegrability. Let I = [a, b] and f :

I R. Then f is not integrable if and only if there exists 0 > 0 such
that for any > 0 there exist tagged partitions P and Q of I such that
< and kQk
< but
kPk
S ( f, Q)|
0 .
|S ( f, P)
Example.
7.2.1
The Squeeze Theorem

Theorem 7.2.2 Let I = [a, b] and f : I R. Then f R(I) is integrable if and only if for any > 0 there exists integrable functions
and in R(I) such that x I
(x) f (x) (x)

and
( ) < .
a
Proof:
1. (I). Suppose f is integrable. Choose = = f .
2. (II). Suppose , R(I) and
Z b
( ) < .
a
if kPk
< , then
3. Then there is > 0 such that P,

Z b
Z b

S (, P)

<
and
< .
S (, P)

a
a
if kQk
< , then
4. Then Q,
Z b
Z b
< S ( f, Q) <
+ .
a
5. Thus
S ( f, Q)|
<
|S ( f, P)
+ 2 < 3 .
a
7.2.2
101
Classes of Riemann Integrable Functions

Definition 7.2.1 Let I R be an interval and : I R. Then is a
step function if its range (I) is a finite set, that is, it takes only a finite
number of values.
The characteristic function I : R R of an interval I is an elementary step function defined by

(
1, x I
(x) =
0, x < I .
An arbitrary step function can be represented as a linear combination of the

elementary step functions.
Lemma 7.2.1 Let a, b, c, d R such that a c < d b. Let I = [a, b]

and J I be a subinterval of I with the endpoints c and d. Then the
characteristic function J of J is integrable on I, J R(I), and
Z b
J = d c .
a
Proof:
1. Isolate the subintervals with the points c and d and get the estimate
(d c)| (tk )|xk xk1 | + (tm )|xm xm1 |
|S ( f, P)
+|c xk | + |d xm1 | < 4 <
by choosing < /4.
Theorem 7.2.3 Let I = [a, b] and : I R be a step function. Then

is integrable on I, R(I).
Proof:
1. By linearity of the integral.

102
Theorem 7.2.4 Let I = [a, b] and f : I R be a continuous function

on I. Then f is integrable on I, f R(I).
Proof:
1. Use the fact that f is uniformly continuous.
2. Define step functions and such that f .
< , then
3. Since f is continuous there is such that if kPk
Z b
( ) < .
a
4. Thus f is integrable.
Theorem 7.2.5 Let I = [a, b] and f : I R be a monotone function

on I. Then f is integrable on I, f R(I).
Proof:
1.
2.
3.
4.
Let f be increasing.
f is bounded.
Define step functions and such that f .
< , then
Since f is increasing, there is > 0 such that if kPk
Z b
( ) [ f (b) f (a)] < ,
a
by choosing .
7.2.3
The Additivity Theorem

Theorem 7.2.6 Additivity Theorem. Let I = [a, b], c I, and f : I
R. Then f is integrable if and only if its restrictions to the subintervals
[a, c] and [c, b] are both integrable.
That is, f R([a, b]) if and only if f R([a, c]) and f R([c, b]).
In this case,
f =
a
f+
a
f.
c
103
Proof: (Read in the textbook).
Corollary 7.2.2 Let I = [a, b] and f : I R. If f is integrable on I

then it is integrable on any subinterval of I.
Proof: Obvious.
Corollary 7.2.3 Let I = [a, b] and f : I R. Let P = {Ik }nk=1 be

a partition of I. If f is integrable on I then it is integrable on every
subinterval Ik for k = 1, . . . , n, and
Z
f =
n Z
X
Ik
Definition 7.2.2 Let I = [a, b] and f : I R be integrable on I. Let

c, d I such that c < d. Then
Z d
Z c
Z c
f =
f,
f = 0.
c
f
k=1
Theorem 7.2.7 Let I be an interval, a, b, c I, and f : I R be

integrable on I. Then
Z b
Z c
Z b
f =
f+
f.
a
The existence of any of these two integrals implies the existence of the
third integral and the above equality.
Proof: (Read in the textbook.)
Examples.
7.2.4
Homework
Exercises: 7.2[7,8,11,12,15]
104
7.3
The Fundamental Theorem
7.3.1
The Fundamental Theorem (First Form)
Definition 7.3.1 Let I = [a, b] and f : I R. The function F : I R

is antiderivative of f if F is differentiable and
F0 = f .
Theorem 7.3.1 The Fundamental Theorem of Calculus (First
Form). Let I = [a, b] and E I be a finite subset of I. Let f, F : I R
be such that
1. F is continuous on I,
2. F 0 (x) = f (x) for all x I \ E,

3. f is integrable on I.
Then
f = F(b) F(a) .
a
Proof:
1. (I). We prove it first for the case E = {a, b}.
2. Let > 0.
if kPk
< , then
3. Then there is > 0 such that for any partition P,

Z b

f < .
S ( f, P)

a
4. Then i {1, . . . , n} there is ui (xi1 , xi ) such that
F(xi ) F(xi1 ) = f (ui )(xi xi1 )
5. Let us choose the tags of the partition P to be ui .
6. Then
= F(b) F(a) .
S ( f, P)
7.3. THE FUNDAMENTAL THEOREM

7. Now we have
105

Z b

F(b) F(a)
a
8. Thus

f < .

F(b) F(a) =
f.
a
9. (II). Now let E = {c1 , . . . , cm }, where

a c1 < c2 < < cm1 < cm b .
10. Then the theorem follows from the additivity of the integral.

Examples.
7.3.2
The Fundamental Theorem (Second Form)

Definition 7.3.2 Let I = [a, b] and f : I R. Let F : I R be
defined for x I by
Z x
F(x) =
f.
a
Then F is the indefinite integral of f with basepoint a.

Theorem 7.3.2 Let I = [a, b] and f : I R be integrable on I and F
be the indefinite integral of f with the basepoint a. Then:
1. F is continuous on I,
2. F is Lipshitz. In particular, if | f (x)| M for all x I, then
x, y I
|F(x) F(y)| M|x y| .
Proof:
1. Let x < y.
Z
2. Then
F(y) = F(x) +
f.
x
106
3. We have
| f (t)|dt M(y x) .
x
4. Thus,
|F(y) F(x)| M|y x| .
Theorem 7.3.3 The Fundamental Theorem of Calculus (Second

Form) Let I = [a, b], c I, f : I R be integrable on I, and F
be the indefinite integral of f with the basepoint a. If f is continuous at
c, then F is differentiable at c and
F 0 (c) = f (c) .
Proof:
1. Let, first, c (a, b).
2. Let > 0.
3. There is > 0 such that if |x c| < , then
| f (x) f (c)| < .
Z
4. We have
F(x) F(c) =
f.
c
5. Then

F(x) F(c) f (c) < .

xc
6. So, as x c,
F 0 (c) = f (c) .
7. For the endpoints we only have to check the left and the right limits.
Theorem 7.3.4 Let I = [a, b], f : I R be integrable on I, and F be

the indefinite integral of f with the basepoint a. If f is continuous on I,
then F is differentiable on I and
F0 = f .
107
Remark. If f is not continuous on I, then the indefinite integral is not equal

to the antiderivative.
Examples.
7.3.3
Substitution Theorem
Theorem 7.3.5 Substitution Theorem. Let I = [a, b] and f : I R
be continuous on I. Let J = [c, d] and g : J R be differentiable on J
and g0 be continuous on J. Suppose g(J) I. Then
Z g(b)
Z d
f (x) dx =
f (g(t)) g0 (t) dt .
g(a)
Proof: Exercise.

Examples.
7.3.4
Lebesgues Integrability Criterion

Definition 7.3.3 Let A R be a set of real numbers. Then A is a null
set if for any > 0 there exists a countable collection of open intervals
{Ik }
k=1 , Ik = (ak , bk ), k N, such that
A
Ik
k=1
and
(bk ak ) .
k=1
Examples.
108
Definition 7.3.4 Let A R and Q(x) be a statement about x A.

We say that Q(x) holds almost everywhere on A (or for almost every
x A), if there exists a null set B such that B A and Q(x) holds for
all x A \ B.
We write in this case
Q(x) for a. e. x A.
Remarks.
Every subset of a null set is a null set.
The union of a countable collection of null sets is a null set.
Every countable set is a null set.
Not every null set is countable!
Examples.
A = Q [0, 1].
Theorem 7.3.6 Lebesgues Integrability Criterion. Let I = [a, b] and

f : I R be a bounded function on I. Then f is integrable if and only
if f is continuous almost everywhere on I.
Proof: Appendix C.
Remarks.
Every step function has finite number of discontinuities, and so is integrable.
Every monotone function has countable number of discontinuities, and so
is integrable.
Dirichlet function is discontinuous at every point, and so is not integrable.
Thomae function is discontinuous at every rational number, and, hence, is
integrable.
Examples.
109
Theorem 7.3.7 Composition Theorem. Let I = [a, b] and f : I R

be integrable on I. Let J = [c, d] and g : J R be continuous on J.
Let f (I) J and h = g f : I R be the composition function. Then
h is integrable on I.
Proof:
1. First, h is continuous at every point where f is continuous.
2. The set Dh of points of discontinuity of h is a subset of the set D f of
points of discontinuity of f .
3. The set Dh is a null set.
4. The function h is integrable on I.
Remarks.
The function f g does not have to be integrable.
If g is not continuous, then h is not necessarily integrable.
Example.
Corollary 7.3.1 Let I = [a, b] and f : I R be integrable on I such

that | f (x)| M for all x I. Then the function | f | is integrable on I and
Z b Z b

f
| f | M(b a) .

a
a
Proof:
1. Let : [M, M] R be defined by (t) = |t|.
2. Then f = | f |.
3. Then | f | is integrable on I.
R
R
R
4. Since | f | | f |, then | f | M(b a).
Theorem 7.3.8 The Product Theorem. Let I = [a, b] and f, g : I R

be integrable on I. Then the product f g is integrable on I.
Proof:
110
1. Let : [M, M] R be defined by (t)t2 .

2. Then f 2 = f is integrable.
3. We have
1
f g = [( f + g)2 f 2 g2 ] .
2
4. Therefore, f g is integrable.
7.3.5
Integration by Parts
Theorem 7.3.9 Integration by Parts. Let I = [a, b] and F, G : I R
be differentiable on I, and F 0 and G0 are integrable on I. Then
Z b
b Z b
0
F G = FGa
FG0 .
a
Proof:
1. We have
(FG)0 = F 0G + FG0 .
2. Thus
Z
FG|ba
FG+
a
FG0 .

Theorem 7.3.10 Taylors Theorem with the Remainder. Let I =
[a, b] and f : I R. Let f (k) , exist on I for k = 1, . . . , n + 1, and
f (n+1) be integrable on I. Then
n
X
f (k) (a)
f (b) =
(b a)k + Rn ,
k!
k=0
where
1
Rn =
n!
f (n+1) (t)(b t)n dt .

a
Proof:

1. Let
1
Rn =
n!
2. Then
Rn =
111
f (n+1) (t)(b t)n dt .

a
1 (n)
f (t)(b t)n |ba + Rn1 .
n!
3. The proof follows by induction.

Examples.
7.3.6
Homework
Exercises: 7.3[7,20,21,22]
112
Chapter 8
Topology of Real Numbers
8.1
Open and Closed Sets in R

Definition 8.1.1
1. Let A R, x A and B (x) = (x , x + ) be the -neighborhood
of x.
2. Any set V x that contains B (x) is called a neighborhood of x.
3. The point x A is an interior point of A if A contains a neighborhood of x.
4. The set A of all interior points of A is the interior of A.

5. The point x A is a boundary point of A if every neighborhood
of x contains points in A and in R \ A.
6. The set A of all boundary points of A i the boundary of A.
7. A is open in R if every point of A is an interior point, i.e. A =
A . In other words, for any x A there is a neighborhood of x
contained in A.
8. A is closed in R if its complement R \ A is open in R.
Examples.
113
CHAPTER 8. TOPOLOGY OF REAL NUMBERS
114
Theorem 8.1.1 Open Set Properties.

1. The union of an arbitrary collection of open sets in R is open in
R. That is, if {A } are open for every , then the set
[
A
is open.
2. The intersection of a finite collection of open sets in R is open in
R. That is, if {Ai }ni=1 are open for every i = 1, . . . , n, then the set
n
\
Ai
i=1
is open.
Theorem 8.1.2 Closed Set Properties.

1. The intersection of an arbitrary collection of closed sets in R is
closed in R. That is, if {A } are closed for every , then
the set
\
A
is closed.
2. The union of a finite collection of closed sets in R is closed in R.
That is, if {Ai }ni=1 are closed for every i = 1, . . . , n, then the set
n
[
Ai
i=1
is closed.
Examples.
8.2. COMPACT SETS
8.1.1
115
The Characterization of Closed Sets
Theorem 8.1.3 Characterization of Closed Sets. Let A R be a

set of real numbers. Then A is closed in R if and only if it contains the
limits of all convergent sequences in A.
Theorem 8.1.4 Let A R be a set of real numbers. Then A is closed in

R if and only if it contains all its cluster points.
8.1.2
8.2
The Characterization of Open Sets

Theorem 8.1.5 Let A R be a set of real numbers. Then A is open in
R if and only if it is the union of countably many disjoint open intervals
in R.
Compact Sets
Definition 8.2.1
Let A R.
1. An open cover of A is a collection of open sets in R whose union

contains A. That is, if U = {U } is an open cover, then
[
U .
A
2. A subcollection of sets from an open cover, which is n open cover

itself, is a subcover.
3. A subcover consisting of finitely many sets is a finite subcover.
4. The set A is compact if every open cover has a finite subcover.
Examples.
Theorem 8.2.1 Let A R be a compact set of real numbers. Then A is

closed and bounded.
Theorem 8.2.2 Heine-Borel Theorem. Let A R be a set of real

numbers. Then A is compact if and only if it is closed and bounded.
CHAPTER 8. TOPOLOGY OF REAL NUMBERS
116
Theorem 8.2.3 Let A R be a set of real numbers. Then A is compact

if and only if every sequence in A has a convergent subsequence whose
limit belongs to A.
8.3
8.3.1
Continuity
8.3.2
Corollary 8.3.1 Let f : R R. Then f is continuous if and only if the

inverse image of every open set is open.
Preservation of Compactness
Theorem 8.3.1 Preservation of Compactness Let A be a compact set

in R and f : A R be continuous on A. Then the range f (A) is
compact.
Theorem 8.3.2 Let A be a compact set in R and f : A R be continuous on A and injective. Then the inverse function f 1 : f (A) R is
continuous on f (A).
8.4
Metric Spaces
8.4.1
Metrics
Definition 8.4.1 Let M be a set and d : M M R be a function
satisfying the following properties:
1. d(x, y) 0 for all x, y M,
2. d(x, y) = 0 if and only if x = y,

3. d(x, y) = d(y, x),
4. d(x, y) d(x, z) + d(z, y) for all x, y, z, M,.
Then the function d is a metric (or the distance function) on M and
the pair (M, d) is a metric space.
8.4. METRIC SPACES
117
Examples.
8.4.2
Neighborhoods and Convergence

Definition 8.4.2 Let (M, d) be a metric space, a M be a point in M
and > 0. The set
B (a) = {x M | d(x, a) < }
is a -neighborhood of a called an open ball of radius with center at

a.
A neighborhood of a is any set that contains an open ball with center at
a.
Definition 8.4.3 Let (M, d) be a metric space, x M, and (xn ) be a

sequence in M. Then the sequence (xn ) converges to x in M if for any
> 0 there is K N such that if n K, then xn B (x).
Examples.
8.4.3
Cauchy Sequences
Definition 8.4.4 Let (M, d) be a metric space, and (xn ) be a sequence

in M. Then the sequence (xn ) is a Cauchy sequence if for any > 0
there is K N such that if n, m K, then d(xn , xm ) < .
Definition 8.4.5 A metric space is complete if it contains the limits

of all Cauchy sequences in it.
Examples.
Bibliography
118
8.4.4
Open Sets and Continuity

Definition 8.4.6
Let (M, d) be a metric space, A M, and x A.
1. The point x A is an interior point of A if A contains a neighborhood of x.

2. The set A of all interior points of A is the interior of A.
3. The point x A is a boundary point of A if every neighborhood

of x contains points in A and in M \ A.
4. The set A of all boundary points of A i the boundary of A.
5. A is open in M if every point of A is an interior point, i.e. A =
A . In other words, for any x A there is a neighborhood of x
contained in A.
6. A is closed in M if its complement M \ A is open in M.
7. A is compact if each open cover of A has a finite subcover.
Definition 8.4.7 Let (M1 , d1 ) and (M2 , d2 ) be metric spaces, c M1 ,

and f : M1 M2 . Then f is continuous at c M1 if for every > 0
there is a > 0 such that
if x B (c), then f (c) B ( f (c)).
Theorem 8.4.1 Global Continuity Theorem.

Let (M1 , d1 ) and
(M2 , d2 ) be metric spaces, and f : M1 M2 . Then f is continuous
on M1 if and only if the inverse image of every open set in M2 is an open
set in M1 .
Theorem 8.4.2 Preservation of Compactness. Let (M, d) be a metric

space and f : M R be continuous on M. Then the range f (M) is
compact in R.
Bibliography
[1] R. G. Bartle and D. R. Sherbert, Introduction to Real Analysis, 3rd Ed.,
John Wiley, 2000
119
120
BIBLIOGRAPHY
Notation
121
122
Notation

Intro Avram

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Intro Avram

Uploaded by

Copyright:

Available Formats

Lecture Notes

Basic Concepts of Analysis

The Real Numbers

2.4.2 The ArchimedeanProperty . . .

3 Sequences and Series

introanal.tex; November 21, 2005; 18:02; p. 1

introanal.tex; November 21, 2005; 18:02; p. 2

The Riemann Integral

Topology of Real Numbers

Sets and Functions

Reading: Appendix A (Logic and Proofs).

The union of the sets A and B is the set

The intersection of the sets A and B is the set

The complement of the set B relative to the set A is the set

Intersection of a collection of sets

1.1. SETS AND FUNCTIONS

Mappings and Functions

Remark. The range of a function is a subset of the codomain:

Definition 1.1.5 A function from a set A to a set B is a set f of ordered

Direct and Inverse Images

The inverse image of a subset H B under f is the subset of A

Special Types of Functions

f is injective (one-to-one) if whenever x1 , x2 then f (x1 ) , f (x2 ).

f is surjective (maps A onto B) if f (A) = B, or R( f ) = B.

1.1. SETS AND FUNCTIONS

Definition 1.1.9 Let f : A B and g : B C be two functions.

(g f )1 (H) = f 1 (g1 (H)) .

Let f : A B and A1 A. The restriction of f to A1 is the function

introanal.tex; November 21, 2005; 18:02; p. 7

Finite and Infinite Sets

Reading: Proofs are in Appendix B.

Empty set has 0 elements.

A set S has n elements if there is a bijection f : Nn S from Nn =

Theorem 1.2.1 Uniqueness Theorem. The number of elements of a

Theorem 1.2.2 The set N is infinite.

2. Let C A. If A has m elements and C has 1 element, then A \ C

1. If A is finite, then B is finite.

A set S is countable if it is either finite or countably infinite.

1.2. FINITE AND INFINITE SETS

Theorem 1.2.5 The set N N is countably infinite.

1. If A is countable, then B is countable.

Theorem 1.2.8 The set Q is countably infinite.

Theorem 1.2.9 The union of a countable collection of countable sets is

Reading: Appendix A, B. Sec. 1.1-1.3.

introanal.tex; November 21, 2005; 18:02; p. 10

The Algebraic and Order Properties of R

R is a field with respect to addition and multiplication.

Definition 2.1.1 Algebraic properties of R. There are two binary

CHAPTER 2. THE REAL NUMBERS

2. 1 R, such that 1 , 0 and

1. Let x, a R and x + a = a. Then x = 0.

2. Let y, b R and yb = b. Then y = 1.

1. Let a, b R, a , 0 and ab = 1. Then b = a1 .

2. Let a, b R and ab = 0. Then either a = 0 or b = 0.

2.1. THE ALGEBRAIC AND ORDER PROPERTIES OF R

Division is defined by: a, b R, b , 0,

Rational and Irrational Numbers

Identify N = Z+ with the subset of R by

Define Z as a subset of R as follows. Identify the zeros

or as the opposite to n. Then

CHAPTER 2. THE REAL NUMBERS

The order Properties of R

The set of negative real numbers

Positive real numbers