Professional Documents
Culture Documents
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 . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
. .
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
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
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
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
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
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
95
95
95
96
97
97
98
98
99
100
101
102
103
104
104
105
107
107
110
111
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
113
113
115
115
115
116
116
116
116
116
117
117
118
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Bibliography
118
Notation
121
introanal.tex; November 21, 2005; 18:02; p. 3
Chapter 1
Preliminaries
1.1
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
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
\
Ak = {x | x Ak for all k N}
k=1
Examples.
introanal.tex; November 21, 2005; 18:02; p. 4
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
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).
R( f ) B
Diagrams.
Function as a relation.
CHAPTER 1. PRELIMINARIES
1.1.4
1.1.5
Let f : A B be a function.
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 ) .
introanal.tex; November 21, 2005; 18:02; p. 6
1.1.7
Composition of Functions
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
1.1.8
Restrictions of Functions
x A1 .
Notation
f1 = f |A1
CHAPTER 1. PRELIMINARIES
1.2
1.2.1
Countable Sets
Definition 1.2.2 A set S is denumerable (countably infinite) if there
is a bijection f : N S .
Examples.
Remark. The union of countably infinite sets is countably infinite.
1. S is a countable set.
2. There is a surjection g : N S .
3. There is an injection f : S N.
1.2.2
Homework
CHAPTER 1. PRELIMINARIES
Chapter 2
The Real Numbers
2.1
(a, b) 7 B(a, b)
3. 0 R such that
0+a=a
4. a R (a) R such that
a + (a) = 0
5. ab = ba
9
10
Definition 2.1.2
1. (ab)c = a(bc)
3. a R, a , 0, a1 R such that
aa1 = 1
4. a(b + c) = ab + ac
Theorem 2.1.1
Theorem 2.1.2
11
2.1.1
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.
introanal.tex; November 21, 2005; 18:02; p. 12
12
2.1.2
1. a + b R+ ,
2. ab R+ ,
3. a R+ exactly one of the following holds:
a R+ ,
(a) R+ ,
a = 0.
0 < R ,
R R+ =
a>0
a0
a<0
a 0.
Let a, b R.
13
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. 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.
14
2.1.3
Inequalities
Examples.
2.2
15
a, if a > 0
0, if a = 0
|a| =
a, if a < 0
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
Examples.
introanal.tex; November 21, 2005; 18:02; p. 16
16
2.2.1
Examples.
2.2.2
Homework
2.3
2.3.1
17
Example.
Definition 2.3.2
inf S
introanal.tex; November 21, 2005; 18:02; p. 18
18
1. s S , s u, and
2. if v < u, then there is s0 S such that v < s0 .
Proof: Exercise.
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
2.3.3
Homework
Exercises: 2.3[6,7,8,9,10]
20
2.4
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.
2.4.2
21
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:
introanal.tex; November 21, 2005; 18:02; p. 22
22
1. Let E = {m N|y < m}
2.4.3
The Existence of
2.4.4
23
2.4.5
Homework
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
2.5. INTERVALS
25
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
introanal.tex; November 21, 2005; 18:02; p. 26
26
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.
introanal.tex; November 21, 2005; 18:02; p. 27
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
Sequence (an ), an S , or f : N S .
Definition 3.1.1
R.
Definition 3.1.2
n N.
29
30
3.1.1
lim xn = x,
xn x
|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).
introanal.tex; November 21, 2005; 18:02; p. 30
31
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| .
3.1.2
Tails of Sequences
32
3.1.3
Examples
3.1.4
Homework
Exercises: 3.1[3(b),8,10,13,14]
3.2
33
Limit Theorems
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
introanal.tex; November 21, 2005; 18:02; p. 33
34
That is n N
|xn | M.
1
2
<
|zn | |z|
lim X
lim Z
Proof:
1. For X Y use triangle inequality.
introanal.tex; November 21, 2005; 18:02; p. 34
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|
36
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.
1
| xn x| |xn x|.
x
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
X is increasing if
xn xn+1 for any n N.
X is decreasing if
39
3.3.1
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
40
Example. Let
1
xn
Show, X is bounded above and increasing, therefore, converging.
x1 = 1,
xn+1 = 2 +
3.3.2
1
x
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,
41
Every point that appears infinitely many times in a sequence is a limit point
of the sequence.
and
3.3.3
Homework
Exercises: 3.3[1,4,5,7,9,10]
42
3.4
k N.
Proof: Obvious.
Example.
or
X 0 (k) = X(A(k)), k N
Example.
introanal.tex; November 21, 2005; 18:02; p. 42
Examples.
3.4.1
44
3.4.2
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:
introanal.tex; November 21, 2005; 18:02; p. 44
3.4.3
Homework
Exercises: 3.4[2,3,5,6,9]
46
3.5
Examples.
Examples.
Definition 3.5.2
A sequence X is contractive if C R such that
0 < C < 1 and n N
47
|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 |
introanal.tex; November 21, 2005; 18:02; p. 47
48
4. By induction
|xn xn1 |
C1
6. As m we get (2).
Examples.
3.5.1
Homework
Exercises: 3.5[2,4,6,8,10]
3.6
49
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.
Examples.
3.7.1
Comparison Tests
Theorem 3.7.4 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.
Examples.
51
CHAPTER 4. LIMITS
52
4.1.1
Notation.
lim f (x) = L,
xc
lim f = L,
xc
xc
f (x) L
Examples.
4.1.2
53
1
n
but | f (xn ) L| 0 .
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
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),
!
f (x)
f
(x) =
,
h
h(x)
p
(x) = (x).
introanal.tex; November 21, 2005; 18:02; p. 53
55
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:
1. Use a sequence X in A such that X , c and lim X = c
introanal.tex; November 21, 2005; 18:02; p. 54
CHAPTER 4. LIMITS
56
xc
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.
4.2.1
Homework
Exercises: 4.2[4,5,11(b,c),12]
Chapter 5
Continuous Functions
5.1
Continuous Functions
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,
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
58
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/.
introanal.tex; November 21, 2005; 18:02; p. 56
59
5.1.1
Homework
Exercises: 5.1[3,7,11,12,13]
60
5.2
61
Examples.
Polynomials.
Rational Functions.
Trigonometric Functions.
5.2.1
Examples.
5.2.2
Homework
Exercises: 5.2[5,6,10,12,13]
62
5.3
Examples.
introanal.tex; November 21, 2005; 18:02; p. 60
5.3.1
63
1
< f (xn ) u
n
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
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.
introanal.tex; November 21, 2005; 18:02; p. 62
65
Remarks.
Closed and bounded intervals are compact sets.
Continuous functions preserve compactness.
Not true for not closed or unbounded intervals.
introanal.tex; November 21, 2005; 18:02; p. 63
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
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.
Examples.
68
1
n
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|
69
1. Easy.
Not all uniformly continuous functions are Lipshitz.
Examples.
5.4.2
70
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
71
5.5.1
(k = 1, . . . , n) .
72
5.6
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.
73
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 .
introanal.tex; November 21, 2005; 18:02; p. 71
74
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
)
1
Dn = x I | j f (x)
,
n
nN
75
Dn
n=1
m X
j f (xk ) f (b) f (a).
n
k=1
m
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.
introanal.tex; November 21, 2005; 18:02; p. 73
76
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
or just
D f.
Remarks.
It is allowed that c is the endpoint of the interval.
77
CHAPTER 6. DIFFERENTIATION
78
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).
79
( 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.
introanal.tex; November 21, 2005; 18:02; p. 76
CHAPTER 6. DIFFERENTIATION
80
6.1.1
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.
introanal.tex; November 21, 2005; 18:02; p. 77
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.
introanal.tex; November 21, 2005; 18:02; p. 78
CHAPTER 6. DIFFERENTIATION
82
6.1.3
Homework
Exercises: 6.1[4,9,10,11(c),13]
6.2
83
f (x) f (c),
( f (x) f (c)).
f (x) f (c)
xc
m = inf f (I) .
2. Then
m 0 M.
introanal.tex; November 21, 2005; 18:02; p. 80
CHAPTER 6. DIFFERENTIATION
84
3. If both m = M = 0 then f = 0.
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)
f (x) f (a)
=0
xa
introanal.tex; November 21, 2005; 18:02; p. 81
85
3. Thus, f is constant.
f (x2 ) f (x1 )
0.
x2 x1
5. Thus, f is increasing.
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).
introanal.tex; November 21, 2005; 18:02; p. 82
CHAPTER 6. DIFFERENTIATION
86
, x,0
x + 2x sin
x
f (x) =
0,
x=0
f (c) f (x)
0
xc
87
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
CHAPTER 6. DIFFERENTIATION
88
1. Let f 0 (c) > 0.
2. We have
f 0 (c) = lim
xc
f (x) f (c)
>0
xc
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.
introanal.tex; November 21, 2005; 18:02; p. 85
6.2.3
89
Homework
Exercises: 6.2[7,10,12,13,17]
CHAPTER 6. DIFFERENTIATION
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 .
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
introanal.tex; November 21, 2005; 18:02; p. 87
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!
CHAPTER 6. DIFFERENTIATION
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.
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 .
introanal.tex; November 21, 2005; 18:02; p. 89
93
1 00
f (c1 )(x1 x0 )2
2
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 6. DIFFERENTIATION
Chapter 7
The Riemann Integral
7.1
7.1.1
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
96
=
S ( f, P)
n
X
k=1
Examples.
7.1.2
Rb
a
f and L2 =
Rb
a
7.1.3
97
Examples
Step functions.
Functions with some isolated discontinuities.
Dirichlet function. (Not integrable).
Thomae function. (Integrable).
7.1.4
kf = k
2. ( f + g) R(I) and
Z
( f + g) =
a
b
a
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
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
99
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
Rb
a
1
.
n
f.
100
Example.
7.2.1
( ) < .
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
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.
102
4. Thus f is integrable.
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
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
n Z
X
Ik
f
k=1
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
7.3.1
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)
introanal.tex; November 21, 2005; 18:02; p. 100
105
Z b
F(b) F(a)
a
8. Thus
f < .
F(b) F(a) =
f.
a
7.3.2
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| .
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.
107
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
Ik
k=1
and
(bk ak ) .
k=1
Examples.
introanal.tex; November 21, 2005; 18:02; p. 103
108
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].
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.
introanal.tex; November 21, 2005; 18:02; p. 104
109
Remarks.
The function f g does not have to be integrable.
If g is not continuous, then h is not necessarily integrable.
Example.
110
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!
Proof:
introanal.tex; November 21, 2005; 18:02; p. 106
111
1 (n)
f (t)(b t)n |ba + Rn1 .
n!
7.3.6
Homework
Exercises: 7.3[7,20,21,22]
112
Chapter 8
Topology of Real Numbers
8.1
Examples.
113
114
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.
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.
introanal.tex; November 21, 2005; 18:02; p. 109
8.1.1
115
8.1.2
8.2
Compact Sets
Definition 8.2.1
Let A R.
116
8.3
Continuous Functions
8.3.1
Continuity
8.3.2
Preservation of Compactness
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,
117
Examples.
8.4.2
Examples.
8.4.3
Cauchy Sequences
Examples.
introanal.tex; November 21, 2005; 18:02; p. 112
Bibliography
118
8.4.4
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