Chapter 1 The Real Numbers: Sets, Sequences, and Fuctions

¤ Sets

Definition. A set E in R is dense in R if for any a, b ∈ R (a < b), there is c ∈ E such that
a < c < b.
Theorem 1. The set of rational numbers Q is dense in R.
Equipotent sets: We call two sets A and B equipotent if there is a one-to-one mapping f
from A and B, denoted by A ∼
= B.

Definition Let Nn = {1, 2, . . . , n} ⊂ N. For any set A, we say:

(1) A set E is finite if E is empty, i.e., E = φ or it has only a finite number of elements in
E. That is, there is a number n for which E is equipotent to Nn .
(2) E is infinite if E is NOT finite.
(3) E is countably infinite if E ∼ = N. Countable sets are sometimes called enumerable or
denumerable sets.
(4) E is (at most) countable if E is finite or countable.
(5) E is uncountable if E is neither finite nor countable.
Theorem 2. The following sets are countable.
(1) N × N × · · · × N
(2) The set of rational numbers Q.
(3) The union of countable collection of countable sets. That is, if {En } be a sequence of
countabe sets, then E = ∪∞
n=1 {En } is countable.

Theorem 3. The sets (a, b), [a, b) and [a, b] are uncountable.

¤ Sequence of Real Numbers

Sequence : A sequence of real numbers is a function on N to R

a : N ∋ n 7−→ a(n) =: an .

Limit point : Let (an ) be a sequence of real numbers. Then, a is called a limit of (an ) i.e.,
lim an = a, if for any ǫ > 0, there exists an integer N > 0 such that
n→∞
|an − a| < ǫ, ∀n ≥ N.
Let a be the limit of (an ). Then, for any ǫ > 0, all but a finite number of an are in the ball
B(x, ǫ).
Theorem 4. (Bolzano-Weirstrass Theorem) Every bounded sequence of real numbers has
a convergent subsequence.
Theorem 5. A subset E of R is closed if and only if it contains all its limit points.
Theorem 6. A monotone sequence converges if and only if it is bounded.
Cauchy Sequence: A sequence (an ) is called Cauchy sequence if for any ǫ > 0, there exists
an integer N > 0 such that
|an − am | < ǫ, ∀n, m ≥ N.
Theorem 7. A sequence of real numbers converges if and only if it is a Cauchy sequence.
Cluster Point: Let (an ) be a sequence of real numbers. Then, a is called a cluster point of
(an ) if for any ǫ > 0 and N > 0, there exists an integer n ≥ N such that
|an − a| < ǫ.
For instance, consider an = (−1)n .
limsup/inf: Let (an ) be a sequence of real numbers. Define

liman = inf sup ak = lim sup an .

n k≥n

Then, ℓ = lim sup an iff (i) For any ǫ > 0, there exists n ∈ N such that ak < ℓ + ǫ for all k ≥ n
and (ii) given ǫ > 0, and n, there exists k ≥ n such that ak > ℓ − ǫ.
Similarly, we can define
liman = sup inf ak = lim inf an .
n k≥n

Properties: Let (an ) be a sequence of real numbers. Then

(1) lim (−an ) = − lim an .
(2) lim an ≤ lim an .
(3) If an → ℓ, then ℓ = lim an = lim an .

¤ Open and Closed Sets of Real Numbers

Properties:
(1) The intersection of any finite collection of open sets is open.
(2) Every open set of real numbers is the union of a countable collection of disjoint open
intervals.
(3) The union of any collection of open sets is open
(4) The union of any finite collection of closed sets is closed.
(5) The intersection of any collection of closed sets is closed.

Theorem 8. (Heine-Borel) Let F be a closed and bounded set in R. Then, each open
covering of F has a finite subcovering.

Continuity: A real-valued function f defined on E is said to continuous at x ∈ E if given

ǫ > 0, there exists δ > 0 such that for all y with |x − y| < δ, we have
|f (x) − f (y)| < ǫ.

Uniform Continuity: A real-valued function f definedon E is said to uniformly continu-

ous (on E) if given ǫ > 0, ∃δ > 0 such that for all x, y with |x − y| < δ, we have
|f (x) − f (y)| < ǫ.
A continuous function on a closed and bounded set is uniformly continuous.
2
Lipschitz Continuity: A real-valued function f defined on E is said to be Lipschitz (on E)
if ∃c > 0 such that for all x, y in E.
|f (x) − f (y)| ≤ c|x − y|
Theorem 9. (Extreme Value Theorem) A continuous function on a closed and boundned
set takes a minimum and a maximum values
Theorem 10. (Intermediate Value Theorem) Let f be a continuous function on [a, b] and
f (a) < c < f (b). Then, there is number x0 ∈ (a, b) such that f (x0 ) = c. set takes a minimum
and a maximum values

Pointwise Convergence: Let (fn ) be a sequence of function defined on E. Then (fn ) is said
to converge pointwise if for every x ∈ E, we have f (x) = lim fn (x) : that is, if given x ∈ E
and ǫ > 0, ∃N such that
|f (x) − fn (x)| < ǫ ∀n ≥ N.
Uniform Convergence: A sequence (fn ) of function defined on E is said to converge uni-
formly if given ǫ > 0, ∃N such that
|f (x) − fn (x)| < ǫ ∀x ∈ E, n ≥ N.

Chapter 2 Lebesgue Measure

2.1 Introduction
Goal: Construct a collection of sets in R called Lebesgue measurable sets and a set function
(say m) of this collection called Lebesgue measurable.
Measure: A measure is a set function (say m) valued in [0, ∞] with the following properties
(pratically):
(1) For each nonempty interval I,
m(I) = length of I = ℓ(I).
(2) m is translation invariant: m(E) = m(E + y).
(3) (countably additivity) If (En ) is a sequence of disjoint sets in A, then
¡[ ¢ X
m En = m(En )
n n
* Ideally, the domain of m is P(R), all subsets of R.

Remark: There is no such measure m on R satisfying the above conditions.

Weakened condition: Construct a set function on a rich class of sets that satisfies the above
three conditions. For this, we take the following two steps:
(1) Construct a set function called outer measure, denoted by m∗ , defined on any set of
R. We will see that outer measure is countably subadditive, i.e.,
¡[ ¢ X
m∗ En ≤ m ∗ (En ).
n n
(2) Determine ‘measurable sets’ and restric m∗
to the collection of measurable sets. Denote
it by m and we will prove that m is countably additive. We call m Lebesgue measure.
3
2.2 Lebesgue Outer Measure

Outer Measure: Let A be a set of real numbers and consider open intervals In with n =
∞
1, 2, · · · . such that A ⊂ ∪ In . The outer measure m∗ (A) is defined by
n=1
∞
X
∗
m (A) = inf ℓ(In ).
A⊂∪In
n=1
Note that
(1) m∗ (φ) = 0.
(2) If A ⊂ B, then m∗ A ≤ m∗ B.
(3) For a single point x, m∗ ({x}) = 0.

Example A countable set has a couter measure zero.

Proposition 1. The outer measure of an interval is its length.
Proposition 2. (Translation invariant) For any set A and a number y, m∗ (A+y) = m∗ (A),
Proposition 3. (Countably subadditivity) Let {Ek } be a countable collection of sets of
real numbers, disjoint or not. Then
∞
∞ X
m∗ ( ∪ Ek ) ≤ m∗ (Ek ).
k=1
k=1

Corollary 1.
(1) If A is a countable set, then m∗ (A) = 0.
(2) The set [0, 1] is not countable.

Remark (Finite subadditivity) If {Ek }nk=1 is a finite collection of sets, disjoint or not. Then
n
n X
m∗ ( ∪ Ek ) ≤ m∗ (Ek ).
k=1
k=1

¤ Some Definitions and Properties

Notations: Let A be a subset in X.

à = {x ∈ X : x 6∈ A}
B ∼ A = {x ∈ X : x ∈ B and x 6∈ A}.

Definition 1. A collection A of subsets of X is called an algebra of sets (or a Boolean algebra)

provided that
(1) If A, B ∈ A, then A ∪ B ∈ A (closed under finite unions).
(2) If A ∈ A, then à ∈ A (closed under complements).

The condition (1) and (2) in the above definition implies that A ∩ B ∈ A (closed under finite
intersections).
4
Proposition 4. Let (Ai ) be a sequence of sets in an algegra A. Then, ∃ (Bi ) of sets in A
such that Bn ∩ Bm = φ for n 6= m and
∞
[ [∞
Ai = Bi .
i=1 i=1

Definition 2. An algebra A of subsets in X is called a σ-algebra (or a Borel field) if (Ai ) is

∞
a sequence of sets, then ∪ An ∈ A (closed under countable unions).
n=1

Proposition 5. Given any collection S of subsets of X, there exists a smallest (σ-) algebra
M containing S.
The smallest (σ-) algebra M containing S is called the (σ-) algebra generated by S.

2.3 The σ-algebra of Lebesgue Measurable sets

Recall that outer measure is defined for any set but it is not countably additive.

Definition 3. A set E is said to be measurable if for any set A ,

m∗ A = m∗ (A ∩ E) + m∗ (A ∩ Ẽ).

Remark:
(1) Since m∗ (A) ≤ m∗ (A ∩ E) + m∗ (A ∩ Ẽ) by proposition 3, we can say
E is measurable iff m∗ A ≥ m∗ (A ∩ E) + m∗ (A ∩ Ẽ).
(2) E is measurable iff Ẽ is measurable.
(3) R and φ are measurable.

Lemma 1. If m∗ (E) = 0, E is measurable. Any countable set is measurable.

Proposition 6. If E1 and E2 are measurable, E1 ∪ E2 is measurable. The union of finite
collection of measurable sets is measurable.
Proposition 7. Let A be a set and E1 , . . . , En be finite sequence of disjoint measurable sets.
Then,
n
£ n ¤¢ X
m∗ A ∩ ∪ Ek = m∗ (A ∩ Ek ).
¡
k=1
k=1

Proposition 8. The union of a countable collection of measurable sets is measurable.

Proposition 9. Every interval is measurable.

FACT: An open set is the countable union of (disjoint) open intervals.

Remark:
(1) Let a and b be real numbers. Then all the types of intervals (a, ∞), (−∞, a], (−∞, b),
[b, ∞) and (a, b) are measurable.
(2) Every borel set is measurable. In particular, each open and cleased set is measurable.
(3) Every Gδ and Fσ sets are measurable.
5
Definition 4. Let S = {all open sets in R}. The smallest σ-algebra B which contains S is
called the collection of Borel sets.
Theorem 11. The family M of measurable sets is a σ-algebra that contains the the σ-algebra
of Borel sets. Each interval, each open set, each closed set, each Gδ set and each Fσ set is
measurable.
Theorem 12. The translate of a measurable set is is measurable.

2.4 Outer and inner approximation of Lebesgue Measurable sets

Remark: Let A and B are measurable sets and A ⊂ B with m(A) < ∞. Then, m(B ∼ A) =
m(B) − m(A).
Proposition 10. Let E be a given set. The followings are equivalent:
(i) E is measurable.
(ii) Given ǫ > 0, there exists an open set O ⊃ E with m∗ (O ∼ E) < ǫ.
(ii∗ ) Given ǫ > 0, there exists a closed set F ⊂ E with m∗ (E ∼ F ) < ǫ.
(iii) There exists a Gδ set G ⊃ E such that m∗ (G ∼ E) = 0.
(iii∗ ) There exists a Fσ set F ⊂ E such that m∗ (E ∼ F ) = 0.
Theorem 13. Let E be a measurable set of finite outer measure. Then for each ǫ > 0, there
n
is a finite disjoint collection of open intervals {Ik }nk=1 such that if O = ∪ Ik , then
k=1
m∗ (E − O) + m∗ (O − E) < ǫ.

2.5 Countable additivity, continuity, and the Borel-Cantelli Lemma

Definition 5. Let E be a measurable set. Define a Lebesgue Measure m by

mE = m∗ E, i.e., m∗ E|M = m.
Proposition 11. Lebesgue measure is countably additive. That is, if (En ) are pairwise disjoint
∞
collection of measurable sets, then ∪ En is measurable and
n=1
∞
∞ X
m( ∪ En ) = m(En ).
n=1
n=1
A countable collection of sets (En ) is said to be ascending if for each k, Ek ⊂ Ek+1 , and
said to be descending if for each k, Ek ⊃ Ek+1 .
Proposition 12. Let (En ) be an ascending sequence of measurable sets. Then the following
properties hold:
(1) If (En ) is an ascending sequence of measurable sets, then
∞
m( ∪ En ) = lim m(En ).
n=1 n→∞
(2) If m(E1 ) < ∞ and (En ) is an decending sequence of measurable sets., then
∞
m( ∩ En ) = lim m(En ).
n=1 n→∞

Let A = {x|a property fails at x}. If m(A) = 0, then the property is said to hold almost
everywhere. (briefly a.e.).
6