Measure and Integration Theory - 4-06-11-2021!16!20-07_Measure and Integration Theory(20MAT22C2) (2)
Uploaded by
mustaphamuhammad281Measure and Integration Theory - 4-06-11-2021!16!20-07_Measure and Integration Theory(20MAT22C2) (2)
Uploaded by
mustaphamuhammad281Master of Science (Mathematics) (DDE)
Semester – II
Paper Code – 20MAT22C2
MEASURE AND
INTEGRATION THEORY
DIRECTORATE OF DISTANCE EDUCATION
MAHARSHI DAYANAND UNIVERSITY, ROHTAK
(A State University established under Haryana Act No. XXV of 1975)
NAAC 'A+’ Grade Accredited University
Material Production
Content Writer: Dr Seema Mehra
Copyright © 2020, Maharshi Dayanand University, ROHTAK
All Rights Reserved. No part of this publication may be reproduced or stored in a retrieval system or transmitted
in any form or by any means; electronic, mechanical, photocopying, recording or otherwise, without the written
permission of the copyright holder.
Maharshi Dayanand University
ROHTAK – 124 001
ISBN :
Price : Rs. 325/-
Publisher: Maharshi Dayanand University Press
Publication Year : 2021
MASTER OF SCIENCE (MATHEMATICS)
Measure and Integration Theory (Semester–II)
Paper code: 20MAT22C2
M. Marks = 100
Term End Examination = 80
Time = 3 hrs Assignment = 20
Course Outcomes
Students would be able to:
CO1 Describe the shortcomings of Riemann integral and benefits of Lebesgue integral.
CO2 Understand the fundamental concept of measure and Lebesgue measure.
CO3 Learn about the differentiation of monotonic function, indefinite integral, use of the fundamental
theorem of calculus.
Section - I
Set functions, Intuitive idea of measure, Elementary properties of measure, Measurable sets and their
fundamental properties. Lebesgue measure of a set of real numbers, Algebra of measurable sets, Borel
set, Equivalent formulation of measurable sets in terms of open, Closed, F and G sets, Non measurable
sets.
Section - II
Measurable functions and their equivalent formulations. Properties of measurable functions.
Approximation of a measurable function by a sequence of simple functions, Measurable functions as
nearly continuous functions, Egoroff theorem, Lusin theorem, Convergence in measure and F. Riesz
theorem. Almost uniform convergence.
Section - III
Shortcomings of Riemann Integral, Lebesgue Integral of a bounded function over a set of finite measure
and its properties. Lebesgue integral as a generalization of Riemann integral, Bounded convergence
theorem, Lebesgue theorem regarding points of discontinuities of Riemann integrable functions,
Integral of non-negative functions, Fatou Lemma, Monotone convergence theorem, General Lebesgue
Integral, Lebesgue convergence theorem.
Section - IV
Vitali covering lemma, Differentiation of monotonic functions, Function of bounded variation and its
representation as difference of monotonic functions, Differentiation of indefinite integral, Fundamental
theorem of calculus, Absolutely continuous functions and their properties.
Note : The question paper of each course will consist of five Sections. Each of the sections I to IV will contain
two questions and the students shall be asked to attempt one question from each. Section-V shall be compulsory
and will contain eight short answer type questions without any internal choice covering the entire syllabus.
Books Recommended :
1. Walter Rudin, Principles of Mathematical Analysis (3rd edition) McGraw-Hill, Kogakusha,
1976, International Student Edition.
2. H.L. Royden, Real Analysis, Macmillan Pub. Co., Inc. 4th Edition, New York, 1993.
3. P. K. Jain and V. P. Gupta, Lebesgue Measure and Integration, New Age International (P)
Limited Published, New Delhi, 1986.
4. G.De Barra, Measure Theory and Integration, Wiley Eastern Ltd., 1981.
5. R.R. Goldberg, Methods of Real Analysis, Oxford & IBH Pub. Co. Pvt. Ltd, 1976.
6. R. G. Bartle, The Elements of Real Analysis, Wiley International Edition, 2011.
Contents
SECTION TITLE OF CHAPTER PAGE NO.
1 MEASURABLE SETS 1-27
2 MEASURABLE FUNCTIONS 28-51
3 LEBESGUE INTEGRATION 52-74
4 FUNCTIONS OF BOUNDED
VARIATIONS 75-86
SECTION – I
MEASURABLE SETS
Introduction
In measure theory, a branch of mathematics, the concept of Lebesgue measure, was given by French
mathematician Henri Lebesgue in 1901. Sets that can be assigned a Lebesgue measure are called
Lebesgue-measurable; the measure of the Lebesgue-measurable set A is here denoted by m*(A).
Lebesgue Measure
In this section we shall define Lebesgue Measure, which is a generalization of the idea of length.
1.1 Definition. The length (I) of an interval I with end points a and b is defined as the difference of
the end points. In symbols, we write.
= – .
1.2 Definition. A function whose domain of definition is a class of sets is called a Set Function. For
example, length is a set function. The domain being the collection of all intervals.
1.3 Definition. An extended real – valued set function defined on a class E of sets is called
Additive if , , = ,
= +
1.4 Definition. An extended real valued set function defined on a class E of sets is called finitely
additive if for every finite disjoint classes , , … . , of sets in E, whose union is also in E,
we have
= ∑
1.5 Definition. An extended real–valued set function defined on a class E of sets is called
countably additive it for every disjoint sequence of sets in E whose union is also in E, we have
∞
∞
= ∑
1.6 Definition. Length of an open set is defined to be the sum of lengths of the open intervals of
which it is composed of. Thus, if G is an open set, then
= ∑
where
= , 1 2 = .
2 Measure and Integration Theory
1.7 Definition. The Lebesgue Outer Measure or simply the outer measure m* of a set A is defined as
∗ = .
where the infimum is taken over all finite or countable collections of intervals {In} such that
Since the lengths are positive numbers, it follows from the definition of ∗ that ∗ .
1.8 Remark: (i) If ⊆ ,then ∗ ≤ ∗ i.e. outer-measure has monotone property.
Proof: By definition of outer-measure, for each ɛ > , there exist a countable collection of open
interval {In} such that ⊆ and
∗ + ɛ > ...1)
now ⊆ ⊆
=> ⊆
∗ ≤
< ∗ + ɛ
∗ < ∗ + ɛ
but ɛ > is arbitrary, ∗ ≤ ∗ hence proved.
(ii) Outer-measure of a set is always non-negative.
1.9 Theorem. Outer measure is translation invariant.
Proof. Let > be given. Then by definition of outer measure, There exist a countable collection of
intervals such that
∗ + > .
Now,
=> + + ,
=> ∗ + + = [length is translation invariant]
∗ +
Since is arbitrary positive number, we have
(2) ∗ + ∗ (1)
To prove reverse inequality, Let > be given. Then by definition of outer measure, There exist a
countable collection of intervals such that
+
∗ + + > .
Now, +
Measurable Sets 3
−
∗ −
∗ < ∗ + +
∗ ∗ + (2)
Then Combining (1) and (2), the required result follows.
i.e., ∗ = ∗ +
1.10 Theorem. The outer measure of an interval is its length.
Proof. CASE (1) Let us suppose, first I is a closed and bounded interval, say I = [a, b]
To prove: m*(I) = ℓ [a, b] =b - a.
Now for each ɛ> 0, I = [a, b] ⊆ (a - ɛ, b+ ɛ) then
by definition of outer-measure
=> m*(I) ≤ ℓ (a - ɛ, b+ ɛ) ≤ (b+ ɛ- a+ ɛ)
=> m*(I) ≤ b-a + 2 ɛ
since ɛ is an arbitrary, m*(I) ≤ b-a = ℓ (I) (1)
Now to prove, m*(I) = b-a, then it is sufficient to prove m*(I) ≥ b-a. let {In} be a countable collection of
open intervals which covering I i.e.
I ⊆
ℓ ≥ b-a for all n ∈ N so it is sufficient to prove that
inf ℓ ≥ b-a
since I = [a, b] is compact, then by Heine Boral theorem, we can select a finite number of open intervals
from this {In} such that their union contains I.
Let the intervals be J1, J2, ..., Jp such that ⊇ [a, b].
Now it is sufficient to prove ℓ ≥ b-a (2)
Now a ∈ I = [a, b], there exist open interval J1 = (a1, b1) from the above-mentioned finite no. of intervals
such that a1 < a≤ b then b1 ∈ I.
Again, there exist an open interval (a2, b2) from the finite collection J1, J2, ..., Jp such that a2 <b1 < b2.
Continuing this, we get a sequence of open intervals
(a1, b1), (a2, b2), ..., (ap ,bp ) from J1, J2, ..., Jp satisfying ai < bi-1 < bi ,i= 2,3,.....,p since the collection is
finite so the process must stop with an interval satisfying ap < bp-1 < bp and ap <b < bp
ℓ ≥ ℓ = ℓ (a1, b1) + ℓ (a2, b2) +.... ℓ (ap, bp)
= (b1 -a1) + (b2 -a2) + ...+ (bp -ap)
= bp + (bp-1 – ap) +...+b1 -a2 – a1
4 Measure and Integration Theory
> bp – a1
> b-a
=> inf ℓ ≥ b-a
=> m*(I) ≥ b-a (4)
Hence result is proved in the case when I closed and bounded interval.
CASE (2) let I be bounded open interval with end points a and b, then for every real no. ɛ> 0 [a+ ɛ, b- ɛ]
⊂ I ⊂ [a, b]
=> m*[a+ ɛ, b- ɛ] ≤ m*(I) ≤ m* [a, b]
=> ℓ [a+ ɛ, b- ɛ] ≤ m*(I) ≤ ℓ [a, b] (by case 1)
=>b- ɛ -a- ɛ≤ m*(I) ≤b-a
since ɛ is arbitrary,
we get b-a ≤ m*(I) ≤ b-a
=> m*(I) =b-a.
CASE (3) if I is the unbounded interval, then for each real no. r> 0, we can find bounded closed interval
J ⊂ I such that ℓ (J)>r
Now J ⊂ I => m* (J) ≤ m* (I)
=> ℓ (J) ≤ m* (I)
=> m* (I) > r since this hold for each real no. r,
we get m* (I) = ∞ = ℓ(I)
i.e. outer-measure is of an interval equal to its length.
1.11 Theorem. Let be a countable collection of sets of real numbers. Then
∗ ≤ ∗ .
Proof. Proof. If one of the sets has infinite outer measure, the inequality holds trivially. So suppose
∗ is finite. Then, given > , there exists a countable collection , } of open intervals such
that , and
, < ∗ +
by the definition of ∗ .
Now the collection , , = , is countable, being the union of a countable number of
countable collections, and covers . Thus
∗ ⋃ , ,
Measurable Sets 5
= ,
< ( ∗ + )
= ∗ +
= ∗ +
= ∗ +
Since is an arbitrary positive number, it follows that
∗ ≤ ∗ .
1.12 Theorem. Outer-measure of singleton set of reals is zero
Proof: Let = Then, since = , ⊆ ( − , + ) ∀ ∈
∗ ≤ ∗ ( − , + )
∗ ≤
≤ ∗ ≤ for each n.
In limiting case ∗ = .
1.13 Theorem. Outer-measure of null set is zero.
Proof: Since ⊆ ( − , ) ∀ ∈
∗ ≤ ∗ ( − , )
∗ ≤
≤ ∗ ≤ for each n. In limiting case ∗ = .
1.14 Corollary. If A is countable, ∗ =
Proof. We know that a countable set is the union of a countable family of singleton. Therefore = ∪
, which yields
∗ = ∗ ∪ ≤ ∗ (by the above theorem)
But as already pointed out outer measure of a singleton is zero. Therefore it follows that
∗
Since outer measure is always a non – negative real number, ∗ = .
1.15 Remark: The Sets , , has outer-measure zero.
6 Measure and Integration Theory
1.16 Remark: Prove that [0, 1] is uncountable.
Proof: Assume on the contrary that the set [0, 1] is countable, then as we know outer-measure of
countable set is zero, then ∗ , = , . , , =
. . , = , a contradiction. therefore [0, 1] is uncountable.
1.17 Corollary. If ∗ = , then ∗ = ∗ .
Proof. Using the above proposition
∗ ≤ ∗ + ∗
= + ∗ (i)
Also
Therefore ∗ ∗ (ii)
From (i) and (ii) it follows that
∗ = ∗
Note:- Because of the property m* ( An ) Σ m* An , the function m* is said to be countably Sub-
additive. It would be much better if m* were also countably additive, that is,
if ∗ ∪ = ∗ .
for every countable collection of disjoint sets of real numbers. If we insist on countable additivity,
we have to restrict the domain of the function m* to some subset m of the set 2R of all subsets of R. The
members of m are called the measurable subsets of R. That is, to do so we suitably reduce the family of
sets on which m* is defined. This is done by using the following definition due to Carathedory.
1.18 Definition. A set E of real numbers is said to be m* measurable, if for every set A R, we have
∗ = ∗ + ∗
Since A = (A E) (A Ec),
It follows from the definition that
∗ = ∗ ∗ + ∗
Hence, the above definition reduces to:
A set E R is measurable if and only if for every set A R, we have
m* A m* (A E) + m* (A Ec).
For example is measurable.
1.19 Theorem. Prove that is measurable set.
Proof: Let A be set of reals, then m* A = m* (A E) + m* (A Ec)
Put E =
Measurable Sets 7
m* (A ) + m* (A c) = m* ( ) + m* (A R)
= 0 + m* A
= m* A
This implies is measurable.
1.20 Theorem. Prove that R is measurable set.
Proof: Let A be set of reals, then
m* A = m* (A E) + m* (A Ec)
Put E = R
m* (A R) + m* (A Rc) = m* (A) + m* (A )
= m* (A) + m* ()
= m* A + 0
= m* A
This implies R is measurable.
1.21 Theorem. If m* E = 0, then E is measurable.
Proof. Let A be any set. Then A E E and so
m* (A E) m* E = 0 (i)
Also A A Ec, and so
m* A m* (A Ec) = m* (A Ec) + m* (A E)
as m* (A E) = 0 by (i)
Hence E is measurable.
1.22 Theorem. Every subset of E is measurable if m* E = 0.
Proof: Let F be any subset of E, where m* E = 0.
then since ⊆
this implies m* F ≤ m* E
this implies m* F ≤ 0
Also m* F ≥ 0
therefore m* F =0.
this implies F is measurable.
1.23 Theorem. Every singleton set is measurable.
Proof: Since outer measure of singleton set is zero and set of measure zero is measurable. Therefore,
singleton set is measurable.
8 Measure and Integration Theory
1.24 Theorem. Every countable set is measurable.
Proof: Since outer measure of countable set is zero and set of measure zero is measurable. Therefore
countable set is measurable.
1.25 Theorem. If a set E is measurable, then so is its complement Ec.
Proof. The definition is symmetrical with respect to Ec, and so if E is measurable, its complement Ec is
also measurable.
1.26 Theorem. Union of two measurable sets is measurable.
Proof. Let E1 and E2 be two measurable sets and let A be any set. Since E2 is measurable, we have
∗ = ∗ + ∗ (i)
and since = (ii)
Therefore by (ii) we have
∗ ∗ + ∗ (iii)
Thus
∗ + ∗
≤ ∗ + ∗ + ∗
= ∗ + ∗
≤ m* A (since E1 is measurable)
i.e. m* (A (E1 E2)) + m* (A (E1 E2)c) m* A
Hence E1 E2 is measurable.
If E1 and E2 are measurable, then E1 E2 is also measurable.
In fact we note that E1, E2 are measurable ⇒ , are measurable ⇒ is measurable ⇒
= is measurable.
Similarly, it can be shown that if E1 and E2 are measurable, then is also measurable.
1.27 Lemma. Difference of two measurable sets is also measurable.
Proof: Let E1 and E2 be two measurable sets. Then is measurable and hence
₁ ∩ = ₁ – ₂ is measurable, being the intersection of two measurable sets.
1.28 Definition. Algebra or Boolean Algebra: - A collection A of subsets of a set X is called an
algebra of sets or a Boolean Algebra if
(i) A, B A => A B A
(ii) A A => Ac A
(iii) For any two members A and B of A, the intersection A B is in A.
Because of De Morgan’s formulae (i) and (ii) are equivalent to (ii) and(iii).
Measurable Sets 9
It follows from the above definition that the collection M of all measurable sets is an algebra. The proof
is an immediate consequence of Theorems 1.25 and 1.26.
1.29 Definition. By a Boolean - algebra or simply a - algebra or Borel field of a collection of sets,
we mean a Boolean Algebra A of the collection of the sets such that union of any countable collection of
members of this collection is a member of A.
From De Morgan’s formula an algebra of sets is a - algebra or Borel field if and only if the
intersection of any countable collection of members of A is a member of A.
1.30 Lemma. Let A be any set, and , , … . , a finite sequence of disjoint measurable sets. Then
∗ = ∗
Proof. We shall prove this lemma by induction on n. The lemma is trivial for
n = 1. Let n > 1 and suppose that the lemma holds for n – 1 measurable sets Ei.
Since En is measurable, we have
∗ = ∗ + ∗ for every set X R.
In particular we may take
= .
Since , , … . , are disjoint, we have
= ∩ = ∩
= =
Hence, we obtain ∗ = ∗ + ∗
(i)
But since the lemma holds for – we have
∗
= ∑ ∗
Therefore (i) reduces to
∗ = ∗ +
∗
= ∗ .
Hence the lemma.
1.31 Lemma. Let A be an algebra of subsets and a sequence of sets in A. Then there
exists a sequence of disjoint members of A such that
∈ = ∈
10 Measure and Integration Theory
Proof. For every i N, let
= − … …
= … …
= … …
Since the complements and intersections of sets in A are in A, we have each . By construction,
we obviously have
Let Dn and Dm be two such sets, and suppose m < n. Then Dm Em, and so
Dm D n E m D n
= … . .
… .
=
….= ……… =
The relation (i) implies ∈
It remains to prove that
∈
For this purpose let x be any member of ∈ . Let n denotes the least natural number satisfying x
En. Then we have
− … … = .
This completes the proof.
1.32 Theorem. The collection M of measurable sets is a - algebra.
Proof. We have proved already that M is an algebra of sets and so we have only to prove that M is
closed with respect to countable union. By the lemma proved above each set E of such countable union
must be the union of a sequence of pairwise disjoint measurable sets. Let A be any set, and let
= . Then En is measurable and . Hence
∗ = ∗ + ∗ ∗ + ∗ .
But, by lemma 1.30,
∗ = ∗ = ∗
Therefore,
∗ ∗ + ∗
Since the left hand side of the inequality is independent of n, we have
∗ ∞ ∗ + ∗
∗ ∞ + ∗ (by countably subadditivity of m*)
= ∗ ∞
+
∗
Measurable Sets 11
= ∗ + ∗ .
which implies that E is measurable. Hence the theorem.
1.33 Lemma. The interval (a, ) is measurable
Proof. Let A be any set and
= ,
= , = − , .
Then we must show that
∗ + ∗ ∗ .
∗ = , then there is nothing to prove. If m* A < , then given > 0 there is a countable
collection {In} of open intervals which cover A and for which
∗ +
Let = , = − , . Then and are intervals (or empty) and =
+ = ∗ + ∗
Since , we have
∗ ∗ ∗ , (iii)
and since, , we have
∗ ∗ ∗ , (iv)
Adding (iii) and (iv) we have
∗ + ∗ ∗ + ∗
= ∗ + ∗
=
≤ ∗ +
But was arbitrary positive number and so we must have ∗ + ∗ ∗ .
1.34 Definition. The collection ß of Borel sets is the smallest - algebra which contains all of the
open sets.
1.35 Theorem. Every Borel set is measurable. In particular each open set and each closed set is
measurable.
Proof. We have already proved that (a, ) is measurable. So we have
, = −, measurable.
∞
Since −, = (−, − ) and we know that countable union of measurable sets is
measurable, therefore (- , b) is also measurable. Hence each open interval,
12 Measure and Integration Theory
, = −, , is measurable, being the intersection of two measurable sets. But each open
set is the union of countable number of open intervals and so must be measurable (The measurability of
closed set follows because complement of each measurable set is measurable).
Let M denote the collection of measurable sets and C the collection of open sets. Then
C M. Hence ß is also a subset of M since it is the smallest - algebra containing C . So each element
of ß is measurable. Hence each Borel set is measurable.
1.36 Definition. If E is a measurable set, then the outer measure of E is called the Lebesgue Measure
of E, is denoted by m. Thus, m is the set function obtained by restricting the set function m* to the
family M of measurable sets. Two important properties of Lebesgue measure are summarized by the
following theorem.
1.37 Theorem. Let be a sequence of measurable sets. Then
≤
If the sets En are pairwise disjoint, then
= .
Proof. The inequality is simply a restatement of the sub-additivity of m*. If {Ei }is a finite sequence of
disjoint measurable sets. So we apply lemma 1.30 replacing A by R. That is , we have
∗ = ∗
∗ = ∑ ∗
and so m is finitely additive..
Let {Ei} be an infinite sequence of pairwise disjoint sequence of measurable sets. Then
∞
And so ⊃
=
∞
∞ ∞
Since the left-hand side of this inequality is independent of n, we have
∞
≥ ∞
The reverse inequality follows from countable sub-additivity and we have
∞
= ∞
Hence the theorem is proved.
1.38 Theorem. Let {En} be an infinite sequence of measurable sets such that ⊂ for each n.
Let m < ∞. Then
∞
m ⋂ = lm m
→∞
Measurable Sets 13
Proof. Let = ∞
and let = − . Then since {En} is a decreasing sequence. We have
= .
Also we know that if A and B are measurable sets then their difference − = is also
measurable. Therefore each Fi is measurable. Thus {Fi} is a sequence of measurable pairwise disjoint
sets.
∞ ∞
ow ⋃ = ⋃ −
∞
= ⋃ ∩
= ∩ ∪
∞
= ∩ ⋂
= ∩
= −
Hence
∞
m ⋃ = m −
∞
⇒ ∑ m = m −
⇒ ∞
m − = m − … (i)
Since = − ∪ , therefore
m = m − + m
⇒ m − m = m − (since m ≤ m < ∞ … (ii)
Again
= − ∪
⇒ m = m − + m
⇒ m − m = m − se ⊂ … (iii)
Therefore (i) reduces to
m − m = ∞
m − m ( us d )
14 Measure and Integration Theory
∞
= lm ∑m − m
→∞
= lm m − m + m − m … − m
→∞
= lm m − m
→∞
= m − lm
→∞
⇒ m = lm m
→∞
⇒ m∞
= lm m
→∞
1.39 Remark. Show that the condition m(E1) <∞ is necessary in the above theorems.
Solution. Let En = [n, ∞)
Then, E1= [ 1,∞)
⟹m(E1) = m[1, ∞) = ∞
We show that the proposition of decreasing sequence does not hold in this case i.e. we want to show that
m(∩∞
) ≠ lm
→∞
Clearly, ⊂ for all n
Now, = [ n, ∞) ⊃ [ n, 2n]
⟹ m( ) ≥ m [n, 2n] = n
⟹ m( ) ≥ n
⟹ lm m = ∞ … (1)
→∞
Now, we claim that m(∩∞ ∞
) = m(∩ , ∞)= 0
For if, ∩∞ ∞
≠ ⟹ there exists x ∈∩
⟹ x ∈ [n, ∞ ) for all n ∈ N
Let x∈ R, so by Archmedian property, we can find a positive integer such that
≤ < +
⟹ x∉ [ + , ∞ ), a contradiction
∴∩∞
=
⟹ m(∩∞
) = 0 … (2)
From (1) and (2), we have
m(∩∞
) ≠
→∞
Measurable Sets 15
S0, theorem does not hold in this case.
1.40 Theorem. Let {En} be an increasing sequence of measurable sets. i.e. a sequence with ⊂
for each n. Let mE1 be finite, then
∞
⋃ = .
→∞
Proof. The sets E1, E2- E1, E3- E2, …, En - En+1 are measurable and are pairwise disjoint . Hence
∪ − ∪ … ∪ − ∪ …
is measurable and
∪ − ∪ … ∪ − ∪ …]
= + ∑ −
= + ∑ −
→∞
But
∪ − ∪ … ∪ − ∪ … is precisely ∞
Moreover,
∑ − = ∑ −
= − + − + ⋯ + −
= −
Thus we have
∞
⋃ = + −
→∞
=
→∞
1.41 Definition : The symmetric difference of the sets A and B is the union of the sets A-B and B-A .
It is denoted by ∆ .
1.42 Theorem. If ∆ = and E1 is measurable, then E2 is measurable. Moreover mE2 =
mE1 .
Proof . We have
E2 = ∪ − − − …(i)
16 Measure and Integration Theory
By hypothesis, both E2 – E1 and E1 – E2 are measurable and have measure zero. Since E1 and E2-E1 are
disjoint, ∪ − is measurable and
∪ − = + = . But, since
− ⊂ ∪ − ,
it follows from (i) that E2 is measurable and
= ∪ − − −
= − = .
This completes the proof.
1.43Definition. Let x and y be real numbers in [0,1]. The sum modulo 1 of x and y , denoted by
+ , is defined by
+ + <
+ =
+ − + ≥
It can be seen that + is a commutative and associative operation which takespair of numbersin [0,1) into
numbers in [0,1).
If we assign to each ∈ , the angle then addition modulo 1 corresponds to the addition of
angles.
If E is a subset of [0,1), we define the translation modulo 1 of E to be the set
+ = [z = x + y for some x ∈ .
If we consider addition modulo 1 as addition of angles, translation module 1 by y corresponds to rotation
through an angle of .
We shall now show that Lebesgue measure is invariant under translation modulo 1.
1.44 Definition. Let x and y be real numbers in [0,1). The sum modulo 1 of x and y, denoted by
+ , is defined by
+ + <
+ =
+ − + ≥
Clearly + ∈ ,
Measurable Sets 17
It can be seen that + is a commutative and associative operation which takes pair of numbers in [0,1)
into numbers in [0,1).
1.45 Definition. If E is a subset of [0,1), we define the translation modulo 1 of E to be the set
+ = [z = x + y for some x ∈ .
We shall now show that Lebesgue measure is invariant under translation modulo 1.
1.46 Lemma. Let ⊂ , be a measurable set. Then for each ∈ , the set + is measurable
and m ( + = .
Proof. Let E1= E∩ , − and E2 = E ∩ − , . Then E1 and E2 are disjoint measurable sets
whose union is E, and so, mE = mE1 + mE2.
we observe that
+ ={ + : x ∈ }
+ + <
= ∈
+ − + ≥ .
But for x ∈ , we have x + y < 1 and so
+ = + , ∈ = + .
and hence + is measurable. Thus
m ( + = + = ,
since m is translation invariant. Also + = + − and so + is measurable and
m( + = . But
+ = + ∪ +
18 Measure and Integration Theory
And the sets ( + and ( + are disjoint measurable sets. Hence + is measurable and
m ( + = + ∪ +
= + + +
= +
= .
This completes the proof of the lemma.
1.47 Theorem: Prove that there exists a non-measurable set in interval [0,1).
Proof: First we define an equivalence relation in the set I= [0,1), By saying that x and y are equivalent
i.e., ∼ if and only if x-y is a rational number.
If x-y is a rational number, we say that x and y are equivalent and write x-y. It is clear that ∼ ; ∼
⟹ ∼ ∼ , ∼ ⟹ ∼ . Thus ‘ ∼’ is an equivalence relation in I.
Hence the relation ∼ partitions the set I = [0,1) into mutually disjoint equivalence classes, that is, classes
such that any two elements of one class differ by a rational number, while any two elements of different
classes differ by an irrational number.
Construct a set P by choosing exactly one element from each equivalence classes. Now we claim that P
is a non-measurable set.
Let < > be a sequence of the rational numbers in [0,1) with r0 = 0 and define Pi = P + ri.
=
(translation modulo 1 of P)
Then P0 = P.
We further prove that (i) ∩ = ∅, ≠ .
(ii) = ,
Proof: (i) Let ∩ ≠ ∅, ≠ .
Let ∈ ∩ . => ∈ ∈
Then∃ , ∈ = +
= +
Measurable Sets 19
+ = +
- = − .
~ .
i.e., ~
=> .
But P has only one element from each equivalence class, therefore we must have = . . , =
But ≠ . Hence a contradiction.
Hence ∩ ≠ ∅, ≠ .
that is, < Pi >is a pair wise disjoint sequence of sets.
(ii) Clearly each ⊂ ,
⊂ , . Let x be any element of [0, 1) = I.
But I is partitioned into equivalent classes therefore x lies in one of the equivalence classes.
x is equivalent to an element say y of P.
x-y is a rational number say ri.
x-y = ri
x = y + ri
= y + .
x ∈ +
x∈
x is in some .
, ⊆ ⋃
, , = ⋃ .
Now we prove P is non-measurable.
Assume that P is measurable, then clearly each Pi is measurable.
And m(Pi) = m +
= m(P) for each i.
Therefore, = = ∞
=
=
>
20 Measure and Integration Theory
But
⋃ = , = , = , .
− .
1.48 Example. The cantor set is uncountable with outer measure zero.
Solution. We already know that cantor set is uncountable. Let Cn denote the union of the closed
intervals left at the nth stage of the construction. We note that Cn consists of closed intervals, each
length . Therefore
∗ ≤ . ∴ ∗ = ∗ ∪ = ∑ ∗
But any point of the cantor set C must be in one of the intervals comprising the union C n, for each n ∈
, and as such ⊂ for all n ∈ . Hence
∗ ≤ ∗ ≤
This being true for each n ∈ , letting → gives ∗ = .
1.49 Example. If E1 and E2 are any measurable sets, show that
∪ + ∩ = + .
Proof. Let A be any set. Since E1 is measurable,
∗ = ∗ ∩ + ∗ ∩ .
We set = ∪ and we have
∗ ∪ = ∗ ∪ ∩ + ∗ ∪ ∩
Adding ∪ to both sides we have
∪ + ∩ = + ∗ ∪ ∩ + ∩ …
But
= ∪ ∩ ∪ ∪ .
Therefore
∪ ∩ ∪ ∪ =
Hence (1) reduces to
∪ + ∩ = + .
Measurable Sets 21
1.50 Theorem. Let E be any set. Then given ∈ > , there is an open set O ⊃ such that ∗ <
∗ + ∈ .
Proof. There exists a countable collection [ In ] of open intervals such that
⊂ and
∞
∑ < ∗ + ∈ .
∞
= ⋃ .
Then O is an open set and
∞
∗ = ∗ ⋃
∞
≤ ∑ m∗
∞
= ∑ l < m∗ +∈ .
1.51 Theorem. Let E be a measurable set. Given ∈> , there is an open set
O ⊃ such that m∗ \ < ∈.
Proof. Suppose first that m E < . Then by the above theorem there is an open set O ⊃ such that
m∗ < m∗ + ∈
Since the sets O and E are measurable, we have
m∗ \ = m∗ − m∗ < ∈.
Consider now the case when m E = . Write the set R of real number as a union of disjoint finite
intervals; that is,
∞
= ⋃ .
Then, if En = E ∩ In, m(En) < . We can, thus, find open sets ⊃ such that
∈
m∗ − < .
∞
Dee = ⋃ . Clel s ope set su tt ⊃ d stses
22 Measure and Integration Theory
∞ ∞ ∞
− = ⋃ − ⋃ ⊂ ⋃ −
∞
∗
m − ≤ ∑ m∗ <∈ .
. :
A set which is countable(finite or infinite) union of closed sets is called an sets. Note: The class of all
sets is denoted by . This F stands for ferme(closed) and for summe(sum).
Example: 1. A closed set.
2. A countable set
3. A countable union of set.
4. An open interval (a, b) since
∞
, = + , − and hence an open set.
- set:
A set which is countable intersection of open sets is a set.
Note: The class of all sets is denoted by This G stands for region and for intersection. The
complement of set is a set and conversely.
Example: 1. An open set in particular an open interval.
2. A closed set
3. A countable intersection of set.
4. A closed interval [a, b] since
, = ∞
− , + .
1.53 Theorem. Let E be any set then
(a) Given ɛ > 0,∃an open set ⊃ such that ∗ < ∗ + ɛ
∃ ⊃ such that m*(E) = m*(G).
Proof: (a) By definition, m* = , where ⊆
if m*(E) = ∞, then clearly result is true. If m*(E) < 0, there is a countable collection {In} of open
intervals such that
⊆ and ∗ + ɛ > (1)
Let = ⊆ , then O is an open set and ⊃
Also ∗ = ∗ ⊆
Measurable Sets 23
≤ ∑ ∗
m*(O) < m*(E) + ɛ [from(1)]
(b) Take ɛ = ∀ ∈ Then by above part, for each ∈ ,∃an open set ⊃ such that
∗ < ∗ +
Now define = ∪∞
, then G is a set.
Also, since each ⊃ E
therefore ∪∞
⊃ this implies ⊃
≥ m*(E) ≤ m*(G) (2)
Also = ∪∞
⊆ ∀
∗ ≤ ∗ for each n
< ∗ + , for each n
in limiting case, we have
m*(G) ≤ m*(E) (3)
Then from (2) and (3), we have
m*(G) = m*(E).
1.54 Theorem. Let E be any set,then the following five statements are equivalent.
(i) E is measurable.
(ii) For given ɛ > 0 , ∃an open set ⊃ such that m*(O – E ) < ɛ
(iii) There exist a set G in with ⊂ , m*(G – E ) = 0
(iv) For given ɛ > 0 , ∃ an closed set ⊂ such that m*(E – F ) < ɛ
(v) There exist a set F in with ⊂ , m*(E – F ) = 0
Proof. Ist we prove (i) implies (ii)
Let E be a measurable set.
Now two cases arrive
Case (i) m*(E) < ∞.
By definition, for given ɛ> 0,there is a countable collection {In} of open intervals such that
⊆ and m*(E) + ɛ > .......(1)
Let O = ⊆ , then O is an open set and ⊃
24 Measure and Integration Theory
Also m*(O) =m*( ⊆ )
m*(O) ≤ ∗
m*(O) < m*(E) + ɛ [from(1)]
m*(O)- m*(E) < ɛ
O = − ∪
m*(O) = m*( − ∪ ) = m*(O-E)+m*(E)
m*(O-E) = m*(O) - m*(E)
⟹ m*(O-E) < ɛ
Case (ii) If m*(E) = ∞
We know that set of real number can be written as countable union of disjoint open intervals
∞
= ⋃
Then = ∩
∞
= ⋂ ⋃
∞
= ⋃ ∩
∞
⟹ = ⋃ , = ∩
clearly each En is measurable and m(En ) is finite.
Because = ∩ ⊆
m*(En ) ≤ l(In) <∞
Then by case (i), for each ∈ , ∃ an open set ⊃ such that
m*(On - En) <
Let us define = ∞
Then O is an open set containing E
Now (O-E) = ∞ ∞ ∞
− ⊂ −
m*(O-E) ≤ m*(∞
− )
≤ ∞ ∗
−
Measurable Sets 25
≤ ∞
= ɛ
⟹ m*(O-E) < ɛ
Now (ii) ⟹ (iii)
Let (ii) holds, then for each ∈ , ∃an open set ⊃ such that
m*(On-E) <
Let us define = ∞
, then G is a set.
Also since each ⊃
therefore ∞
⊃ this implies ⊃
G-E = ∞
− ⊆ −
m*(G-E) ≤ m*(On-E) <
Since n is arbitrary
m*(G-E) ≤ 0
⟹ m*(G-E) = 0.
Now (iii) ⟹ (i)
Let (iii) holds, then for given set E, ∃ a set ⊃ such that m*(G-E) = 0
⟹ G - E is measurable.
Now E = G- (G-E)
Now E is measurable being difference of two measurable sets.
Thus (i) ⟺ (ii) ⟺ (iii)
Now to show (i) ⟹ (iv)
Let (i) holds, and ɛ > 0 be given
then by (ii), for given set Ec, ∃ an open set ⊃ such that m*(G - Ec) < ɛ
Since ⊃ ⟹ ⊆
Let F = Gc
then F is a closed set contained in E,
Now E-F = ∩ = EՈG = GՈE = G – Ec
Now m*(E-F) = m*(G-Ec) < ɛ
m*(E-F) < ɛ.
To Show (iv) ⟹ (v)
Let (iv) holds, then for each ∈ ,∃a closed set ⊂ such that
26 Measure and Integration Theory
m*(E – Fn ) <
Let us define = ∞
Then F is a set.
Also, since each ⊂ ⟹ ∞
⟹ ⊆
Now E – F = E - ∞
⊆ −
⟹ m*(E-F) ≤ m*(E – Fn ) <
⟹ m*(E-F) ≤
Since n is arbitrary.
m*(E-F) ≤ 0
⟹ m*(E-F) = 0.
Now (v) ⟹ (i)
Let (v) holds, then for general E, ∃ a set F such that m*(E-F) = 0
⟹ E-F is measurable.
= − ∪
⟹ E is measurable.
This completes the proof.
(b) Take ɛ = ∀ ∈
Then by above part, for each ∈ , ∃ an open set ⊃ such that
m*(On) < m*(E) +
∞
Now define = , then G is a set.
Also since each ⊃
therefore∞
⊃
this implies ⊃
⟹ m*(E) ≤ m*(G) ......(2)
Also G = ∞
⊆ ∀
m*(G) ≤ m*(On) for each n < m*(E) + , for each n in limiting case, we have
m*(G) ≤ m*(E)...(3)
Then from (2) and (3), we have
m*(G) = m*(E).
Measurable Sets 27
1.55 Theorem. Let E be a set with m* E < . Then E is measurable iff given > 0 , there is a finite
union B of open intervals such that m*(E B) <
Proof. Suppose E is measurable and let > 0 be given. The (as already shown) there exists an open set
O E such that m* (O- E) < . As m*E is finite, so is m*O. Since the open set O can be written as the
2
union of countable (disjoint) open intervals {Ii}, there exists an n N such that
l(I ) 2
i n 1
i (In fact m* O = l(I )
i n 1
i l(I ) 2
i n 1
i because m* O < )
Set B = . . Then E B = (E B) (B -E) (O - B) (O -E) . Hence
m*(E B) m* + m*(O-E) + = .
Conversely, assume that for a given > 0 , there exists a finite union B = . if open intervals with
m* (E B) < . Then using “Let be any set. The given > 0 there exists an open set O E such
that m* O < m* E + there is an open set O E such that
m* O < m* E + (i)
If we can show that m* (O E) is arbitrary small, then the result will follow from “Let E be
set. Then the following are equivalent (i) E is measurable and (ii) given > 0 there is an open set O E
such that m * (O E) < ”. Write S = ∩ . Then S B and so
S E = (E - S) (S - E) (E S) (B E) . However,
E \ S = ( E OC ) ( E BC ) = E B, because E O . Therefore
S E (E B) (B E) = E B , and as such m* (S E) < . However,
E S (S E)
and so m* E < m* S + m* (S E)
<m*S+ ii)
Also,
O E = (O S) (S E)
Therefore
m* (O \ E) < m* O m* S +
< m* E + m* S + (using(i))
< m* S + + m*S + using(ii))
< m*S + + m* S +
=3.
Hence E is measurable.
SECTION – II
MEASURABLE FUNCTIONS
Measurable Function: An extended real valued function f defined on a measurable set E is said to be
measurable function if {x| f(x) > } is measurable for each real number .
2.1 Theorem. A constant function with a measurable domain is measurable.
Proof: Let f be a constant function with a measurable domain E and Let f : E→ be a constant function
i.e., f(x) = k ∀ ∈ .
To show that {x| f(x) > } is measurable for each real number .
, >
{x| f(x) > } = , =
, <
Since both are measurable, it follows that the set {x| f(x) > } and hence f is measurable.
2.2 Theorem. Let f be an extended real valued function defined on a measurable set E, Then f is
said to be measurable (Lebesgue function) if for any real any one of the following four
conditions is satisfied.
(a) {x| f(x) > } is measurable
(b) {x| f(x) ≥ } is measurable
(c) {x| f(x) < } is measurable
(d) {x| f(x) ≤ } is measurable.
Proof: We show that these four conditions are equivalent. First of all we show that (a) and (b) are
equivalent. Since
{x| f(x) > } = {x| f(x) ≤ }c
And also we know that complement of a measurable set is measurable, therefore (a) ⟹(d) and
conversely.
Similarly since (b) and (c) are complement of each other, (c) is measurable if (b) is measurable and
conversely.
Therefore, it is sufficient to prove that (a) ⟹(b) and conversely.
Firstly we show that (b) ⟹ .
The set {x| f(x) ≥ } is given to be measurable.
Now
Measurable Functions 29
{x| f(x) > } = {x| f(x) ≥ + }
n 1
But by (b), {x| f(x) ≥ + } is measurable. Also we know that countable union of measurable sets is
measurable. Hence {x| f(x) > } is measurable which implies that (b) ⟹(a).
Conversely, let (a) holds. We have
{x| f(x) ≥ } = {x| f(x) ≥ − }
n 1
The set {x| f(x) > − } is measurable by (a). Moreover, intersection of measurable sets is also
measurable . Hence {x| f(x) ≥ } is also measurable . Thus (a) ⟹ (b).
Hence the four conditions are equivalent.
2.3 Remark: We can say that f is measurable function if for any real number , any of the four
conditions in the above theorem holds.
2.4 Lemma. If is an extended real number then these four conditions imply that {x| f(x) = } is
also measurable.
Proof. Let be a real number, then
{x| f(x) = } = {x| f(x) ≥ } ∩ {x| f(x) ≤ }.
Since {x| f(x) ≥ } and {x| f(x) ≤ } are measurable by conditions (b) and (d), the set {x| f(x) = } is
measurable being the intersection of measurable sets.
Suppose = . Then {x| f(x) = } = {x| f(x) >
n 1
Which is measurable by the condition (a) and the fact intersection of measurable sets is measurable.
Similarly when = −, then
{x| f(x) = −} = {x| f(x) < −, which is again measurable by conditions (c).Hence the
n 1
results follows.
2.5 Theorem: If f is measurable function on each of the sets in a countable collection {E i} of
disjoint measurable sets, then f is measurable on E = .
Proof: Let E = . Then E is measurable being countable union of measurable sets is measurable.
Let be any real number.
Consider the set {x∈ | f(x) > } = ∈ : > is measurable.
Because f is measurable on each Ei.
∈ : > is measurable.
30 Measure and Integration Theory
{x∈ | f(x) > } is measurable.
Hence f is measurable on E.
2.6 Theorem: If f is measurable function on E and ⊆ is measurable set then f is a measurable
function on E1.
Proof: Let be any real number.
Consider the set {x∈ | f(x) > } = {x∈ | f(x) > }∩ is measurable.
2.7 Theorem. If f and g are measurable functions on a common domain E, then the set
A= {x∈ E: f(x)< g(x)} is measurable.
Proof. For each rational number r, define
= { x∈ E: f(x) < r < g(x)}
Or we can write
= { x∈ E: f(x)< r}∩ {x∈ E: g(x)> r}
Since f and g are measurable on E, so the two sets on R.H.S. are measurable sets is measurable.
Now, we observe that
{x∈E: f(x) <g(x)} = ∈
Since the rationals are countable, so A is countable union of measurable sets and so is measurable.
This proves the theorem.
2.8 Theorem. A continuous function defined on a measurable set is measurable.
Proof. Let f be a continuous function defined on measurable set E. Let be any real number. We now
claim that {x ∈ E : f(x) ≥ } is closed.
Let A= { x ∈ E: f(x) ≥ } (1)
To prove that A is closed, it is sufficient to show that ⊆ . (2)
being derived set of A.
Let ∈ be arbitrary element. Then ∈ implies is limit point of A.
It implies that there exist a sequence { } whose elements ∈ such that
lm =
→∞
Moreover, f is continuous at ; it follows that by definition of continuity → implies )→
(3)
By (2), we see that ∈ A for all n ∈ N.
⇒ f( ) ≥ for all n∈ N.
⇒ lm ≥
→∞
Measurable Functions 31
⇒ ≥ by virtue of (3)
⇒ ∈ by (1)
Further any ∈ implies ∈
⇒ ⊆
⇒ A is closed
⇒A is measurable
⇒{x ∈ E: f(x) ≥ } is measurable.
Hence a continuous function f is measurable on E.
Converse of above theorem is not true, that is, A measurable function need not be continuous.
2.9 Example. Consider a function f: R → [0,1]
≤ ≤
defined by f(x) = .
Clearly function is not continuous since 0 is the point of discontinuity.
For any real number ,
, ≥
{x ∈N: f(x) > }= , <
,, ≤ <
Since R, , [0,1) are measurable implies f is measurable function on E.
2.10 Theorem. Let f be a function defined on a measurable set E then f is measurable iff for any
open set G in R the inverse image f-1 (G) is measurable set.
Proof. Let f be a measurable function and let G be any set in R. Since every open sets can be written
as countable union of disjoint open intervals.
Suppose, G = = ,
Then f-1 (G) = f-1( ) =
= : ∈
= ∶ ∈ ,
but {x : f(x) ∈ , } = {x : < < }
= {x : > }∩ {x : < }
Since f is measurable function. So both sets on R.H.S. are measurable and hence
{x : f(x) ∈ , } is measurable.
Again f-1 (G) is measurable. [ since countable union of measurable sets is measurable ]
32 Measure and Integration Theory
Conversely:- Let f-1 (G) be measurable for every open set G in R. We have to prove that f is
measurable function.
Take G = (, ) where is any real no.
Then f-1(, ) is measurable
that is, {x : f(x) ∈ (, )} is measurable
that is , {x : f(x) > } is measurable.
Thus f is measurable function.
2.11 Theorem. Let f be continuous and g be measurable function then fog is measurable.
Proof. Let be any real number then
{x : fog(x) > } = {x : f(g(x)) > }
= {x : f(g(x)) ∈ (, )}
= {x : g(x) ∈ f-1(, )}.
Now, (, ) is open subset of R and f is continuous implies f-1(, ) is open set.
Hence, it can be written as countable union of disjoint open intervals say
f-1(, ) = = , .
Therefore,
{x : fog(x) > } = {x : g(x) ∈ } = g -1
=
= ∶ ∈
= ∶ ∈ ,
= ∶ < <
= ∶ > ∩ ∶ < .
Since g is measurable function. Both sets on R.H.S. are measurable and their intersection is measurable.
Also countable union of measurable sets is measurable. Hence the result.
2.12 Definition. A function f is said to be a step function iff
f(x) = Ci , < x < for some subdivision of [a, b] and some constants Ci .
, ≤ ≤
Example: A function f : [0, 1] → defined as f(x) = where , are constant, f is a step
, ≤ ≤
function.
Remark: Every step function is a measurable function.
Measurable Functions 33
2.13 Theorem. For any real no c and two measurable real- valued functions, f and g, the functions
f +c, cf, f+g, f-g, fg and f/g (g≠ ), are all measurable.
Proof. We are given that f is measurable function and c ais any real number. Then for any real number
{x| f(x) + > } = {x| f(x) > − }
But {x| f(x) > − } is measurable by the condition (a) of the definition. Hence
{x| f(x) + > } and so | f(x) + is measurable.we next consider the function cf . in case c= , cf is
the constant function 0 and hence is measurable since every constant function is continuous and so
measurable. In case c > we have
α
{x| cf(x) > = {x| f(x) > }, and so measurable.
In case c < , we have {x| cf(x) > ={x| f(x)< } and so measurable.
Now if f and g are two measurable real valued functions defined on the same domain, we shall show that
f+g is measurable. To show that it is sufficient to show that the set
{x| f(x)+ > } is measurable.
if f(x) +g(x)> , then | f(x) > − and by he cor. of the axiom of Archimedes there is a rational
number r such that − <r<f(x)
since the functions f and g are measurable , the sets {x| f(x) > and {x| f(x) > − } are measurable.
Therefore, there intersection Sr={x| f(x) > − }∩ {x| f(x) > − } also measurable.
It can be shown that {x|f(x)+ > =∪ {Sr | r is rational}
Since the set of rational is countable and countable union of measurable sets is measurable , the set
∪ {Sr | r is rational} and hence {x|f(x)+ > is measurable which proves that
f(x) +g(x) is measurable. From this part it follows that f- g = f (-g) is also measurable, since when g is
measurable (-g) is also measurable. Next we consider fg.
1
The measurability of fg follows that from the identity fg = [( f g ) 2 f 2 g 2 ] , if we prove that f2
2
is measurable when f is measurable. For this it is sufficient to prove that
{x|∈ f2(x) > }, is real number, is measurable.
Let be a negative real number. Then it is clear that the set {xf2(x) > } = E (domain of the
measurable function f). But E is measurable by the definition of f. Hence {xf2(x) > } is measurable
when < .
Now let ≥ , then {xf2(x) > }= {x| f(x) > √} ∪{x| f(x) < −√}.
Since f is measurable, it follows from this equality that {xf2(x) > } is measurable for ≥ .
Hence f2 is also measurable when f is also measurable. Therefore, the theorem follows from the above
identity, since measurability of f and g imply the measurability of f+g.
34 Measure and Integration Theory
Consider (g≠ = f.
First we have to prove that is measurable.
Consider the set ∶ ( ) > = ∶ >
: > =
: > ∩ : < >
=
: > ∪ : > ∩ : < <
Since g is measurable in each case ,i.e., ∶ ( ) > is measurable.
is measurable.
Since f and are measurable.
is measurable.
Now If f is measurable then |f| is also measurable.
It suffices to prove that measurability of the set {x| f(x) > }= E (domain of f)
But E is assumed to be measurable. Hence {x| f(x) > }={x| f(x) > }∪{x| f(x)< − }
The right hand side of the equality is measurable since f is measurable. Hence {x| f(x) > }is
measurable. Hence |f| is measurable.
2.14 Remark: Converse of (vii) is not true.
Example: Let P be a non-measurable subset of [0, 1) = E
Define a function f : E→ R as
∈
f(x) =
−
f is not measurable because : > = : = = − .
, ≥
Also, for any real , : > = : > =
, <
Since E and are measurable.
: > .
2.15 Theorem. Let { fn }n1 be a sequence of measurable functions. Then sup{f 1,f2,…,fn}, inf
{f1,f2,…,fn}, sup,inf , lim f n and lim f n are measurable.
n n n
Proof. Define a function M(x) = , , … , we shall show that
{x| M(x)> }is measurable. In fact {x| M(x)> }= : >
Measurable Functions 35
Since each fi is measurable, each of the set {x| fi(x)> }is measurable and therefore their union is also
measurable. Hence {x| M(x)> } and so M(x) is measurable. Similarly we define the function m(x) =
inf {f1,f2,,,,fn}, since {x| m(x)< = : < and
since {x| fi(x)< }is measurable on account of the the measurability of fi, it follows that {x| m(x)< }
and so m(x) is measurable . Define a function M’(x) = sup f n ( x) = sup{f1,f2,,,,fn}
n
We shall show that the set {x| M’(x)> } is measurable for any real .
Now {x| M’(x)> }= ∞
: > is measurable, since each fn is measurable.
Similarly if we define m’(x) = inf f n ( x) , then {x| m’(x)< }= ∞
: < and therefore
n
=
measurability of fn implies that of m’(x). Now since and
≥
= , the upper and lower limit are measurable.
≥
2.16 Corollary: If {fn} is a sequence of measurable functions converging to f. Then f is also
measurable.
Proof: Since{fn} converges to f ,i.e., lm =
→∞
= = lm
Then
→∞
=
i.e., f =
.
Hence f is measurable because
2.17 Corollary: The set of points on which a sequence {fn} of measurable functions converges is
measurable.
.
Proof: By above theorem
− .
, : − = ∀.
, =
: − = .
.e.,
: = .
.e., set of these points for which{fn}converges is measurable.
2.18 Definition. Let f and g be measurable functions. Then we define
= Max (f, 0)
= Max (-f, 0)
f ∨g = i.e. Max (f, g) and
f ∧g = i.e. min (f, g)
36 Measure and Integration Theory
2.19 Theorem. Let f be a measurable function. Then f and are both measurable.
Proof. Let us suppose that f > 0 . Then we have
f = f − (i)
Now let us take f to be negative.
Then f = Max (f, 0) = 0, (ii)
f = Max (-f, 0) = -f
Therefore on subtraction f = f −
In casef = 0 , then f = 0, = 0. (iii)
Therefore f = f −
Thus for all f we have, f = f −
Also adding the components of (i) we have
f = |f| = f + (v)
since f is positive. And from (ii) when f is negative we have
f + = 0 - = = |f| (vi)
In case f is zero, then
f + = 0 + 0 = 0 = |f| (vii)
That is for all f, we have
|f| = f + (viii)
Adding (iv) and (viii) we have f + |f| = 2 f ,
f = (f + |f| ) (ix)
Similarly on subtracting we obtain f = (f − |f| ) (x)
Since measurability of f implies the measurability of | f | it is obvious from (ix) and (x) that f and f are
measurable.
Measurable Functions 37
2.20 Theorem. If f and g are two measurable functions, then f ∨g and f ∧g are measurable.
Proof. We know that
f∨g =
f ∧g =
Now measurability of f _ measurability of |f|. Also if f and g are measurable, then f+g, f-g are
measurable. Hence f∨g and f ∧g are measurable.
1, x E
2.21 Definition. Characteristic function of a set E is defined by E ( X )
0, x E
This is also known as indicator function.
2.22 Examples of measurable function
1, x E
Example. Let E be a set of rationals in [0,1]. Then the characteristic function E ( X ) is
0, x E
measurable.
1, x E
Proof. For the set of rationals in the given interval, we have E ( X )
0, x E
It is sufficient to prove that { x | E ( X ) > is measurable for any real .
Let us suppose first ≥ . Then { x | E ( X ) > ={ x | E ( X ) >
Hence the set { x | E ( X ) > is empty in this very case. But outer measure of any empty set is zero.
Hence for ≥ , the set { x | E ( X ) > and so E ( X ) is measurable .
Further let 0≤ ≤ . Then { x | E ( X ) > = E
But E is countable and therefore measurable. Hence E ( X ) is measurable.
Lastly, let ≤ . Then { x | E ( X ) > } = [0,1] and therefore measurable. Hence the result.
2.23 Theorem. Characteristic function A is measurable if and only if A is measurable.
1, x A
Proof. Let A be measurable. Then A ( x)
0, x A
Hence it is clear from the definition that domain of A is A ∪Ac which is measurable due to the
measurability of A. Therefore, if we prove that the
38 Measure and Integration Theory
set {x | A( x ) > } is measurable for any real , we are through.
Let ≥0 . Then {x | A( x ) > }= {x | A( x ) = }=A(by the definition of Characteristic function.)
But A is given to be measurable. Hence for ≥0. The set {x | A( x ) > } is measurable.
Now let us take < 0 . Then {x | A( x ) > }= A∪AC
Hence {x | A( x ) > } is measurable for < 0 also, since A ∪AC has been proved to be measurable.
Hence if A is measurable, then A is also measurable. Conversely, let us suppose that A( x ) is
measurable. That is,
the set {x | A( x ) > } is measurable for any real .
Let ≥0 . Then {x | A( x ) > }= {x | A( x ) = }= A
Therefore, measurability of{x | A( x ) > } implies that of the set A for ≥0 . Now consider < 0. Then
{x | A( x ) > }= A∪AC
Thus measurability A( x ) of implies measurability of the set A∪AC which imply A is measurable.
2.24 Simple Function: Let f be a real valued function defined on X. If the range of f is finite. We say
that f is a simple function.
1, x E
Let E⊆X and put E ( X )
0, x E
Suppose the range of f consists of the distinct number c1, c2, …, cn.
Let Ei = : = = , , … ,
Then f =
i.e., every simple function is a finite linear combination of characteristic function. It is clear that f is
measurable if and only if the sets E1, E2, …, En are measurable.
2.25 Remarks:
1. Every step function is a simple function.
2. Every simple function is measurable.
Proof: Let f be a simple function defined as above.
Then we have
f(x) = (x)
= 1 + 2 + ⋯ +
∴ = , ∈
= , ∈
∴ = , ∈
Measurable Functions 39
∴ = : =
Since each Ei is measurable. Thus is measurable because is measurable if and only if A is
measurable.
Hence f is measurable.
3. Characteristic function of measurable set is a simple function.
4. Product of the simple function and finite linear combination of simple functions is again a
simple function.
2.26 Theorem. (Approximation Theorem). For every non-negative measurable function f, there
exists a non-negative non-decreasing sequence {fn } of simple functions such that = ,
→
x ∈E
In the general case if we do not assume non-negativeness of f , then we say For every measurable
function f, there exists a sequence {fn }, n ∈N of simple function which converges (pointwise) to f . i.e.
“Every measurable function can be approximated by a sequence of simple functions.”
Proof. Let us assume that f(x) ≥0 and x ∈E . Construct a sequence
, ≤ < = , ,
fn(x) = } for every n ∈N.
, ≥
If we take n = 1, then
, ≤ < = ,
f1 (x) = }
, ≥
, ≤ <
That is, f1(x) = ,
≤ <
≥
Similarly taking n = 2, we obtain
, ≤ < = ,, , ,8
f2(x) = 4 4 4 }
, ≥
That is,
≤ < 4
≤ <
4 4
f2(x)= … … … … … … …
………………..
7 7 ≤ <
4 4
≥
40 Measure and Integration Theory
Similarly we can write f3(x) (x) and so on. Clearly all fn are positive whenever f is positive and also it is
clear that fn ≤ fn+1. Moreover fn takes only a finite number of values. Therefore {fn} is a sequence of
non-negative, non decreasing functions which assume only a finite number of values.
Let us denote
Eni = , = ∈ ≤ <
and
En = . = ∈ ≥
Both of them are measurable. Let
n 2n
i 1
fn n
Eni n En for every n ∈ N .
i 1 2
n 2n
i 1
Now 2
i 1
n
E is measurable, since E
ni ni
has been shown to be measurable and characteristic
function of a measurable set is measurable. Similarly En is also measurable since En is
measurable. Hence each fn is measurable. Now we prove the convergence of this sequence.
Let f(x) < . That is f is bounded. Then for some n we have
≤ <
- ≤ - <
0 ≤f(x) -
<
0 ≤f(x) – fn(x) <
(by the def of fn (x))
f(x) ≤ fn (x) <
or − ≤ < ∀ ≥ ∈ .
since m does not depend upon point.
Therefore, convergence is uniform.
Let us suppose now that f is not bounded. Then f(x) =
f(x) ≥ ∈
But fn (x) = n
lm = = .
→∞
Measurable Functions 41
When we do not assume non-negativeness of the function then since we know that f and f
are both non-negative, we have by what we have proved above
f lim n '( x) (i)
n
f lim n ''( x) (ii)
n
where ∅ and ∅ are simple functions. Also we have proved already that
f = f −
Now from (i) and (ii) we have
f − = lim n '( x) − lim n ''( x)
n n
= lim(n '( x) n ''( x))
n
= lim n ( x)
n
(since the difference of two simple functions is again a simple function). Hence the theorem.
We now introduce the terminology “almost everywhere” which will be frequently used in the Sequel.
2.27 Definition. A statement is said to hold almost everywhere in E if and only if it holds
everywhere in E except possibly at a subset D of measure zero.
(a) Two functions f and g defined on E are said to be equal almost everywhere in E iff
f(x) =g(x) everywhere except a subset D of E of measure zero.
(b) A function defined on E is said to be continuous almost everywhere in E if and only
if ther
2.28 Theorem. (a) If f is a measurable function on the set E and ⊆ is measured set, then f is a
measurable function on .
(b) If f is a measurable function on each of the sets in a countable collection{Ei } of disjoint measurable
sets, then f is measurable.
Proof. (a) For any real , we have {x ∈ , f(x) > } = { x ∈E; f(x) > } ∩E1. The result follows as
the set on the right-hand side is measurable.
(b)Write E = Ei , Clearly, E, being the union of measurable set is measurable. The result now
i 1
follows, since for each real , we have
E = { x ∈E: f(x) > } = { x ∈ Ei : f(x) > }
i 1
42 Measure and Integration Theory
2.29 Theorem. Let f and g be any two functions which are equal almost everywhere in E. If f is
measurable so is g.
Proof. Since f is measurable, for any real, the set {x | f(x) > } is measurable. We shall show that the set
{x | g(x) > } is measurable. To do so we put
E1 = {x | f(x) > } and E2 = {x | g(x) > }. Consider the sets
E1 – E2 and E2 – E1.
∵ ∈ − ⟹ ∈ ∉
These are subsets of : ≠
> , ≯ ⟹ ≠
But f = g a.e.
m: ≠ = 0
− ⊆ : ≠
m( − ) ≤ m: ≠ = 0
m( − ) ≤ 0 But m( − ) ≥ 0
m( − ) = 0
Similarly m( − ) =0
∴ m( − ) = 0 = m( − )
( − ) and ( − ) are measurable.
E2 = [E1 ∪( E2 – E1)] - ( E1 – E2)
Since E1, E2 – E1 and ( E1 – E2)C are measurable therefore it follows that E2 is measurable. Hence the
theorem is proved.
2.30. Corollary. Let {fn} be a sequence of measurable functions such that lim f n f almost
n
everywhere. Then f is a measurable function.
Proof. We have already proved that if {fn} is a sequence of measurable functions then lim f n is
n
measurable. Also, it is given that lim f n = f a.e. Therefore, using the above theorem, it follows that f is
n
measurable.
2.31 Definition: (Restriction of f to E1)
Let f be a function defined on E, then the function f1 defined on E1 contained in E .i.e., E1 ⊆ E by f1(x) =
f(x), x ∈ E1 is called restriction of f to E1 and denoted by f/ E1.
2.32 Exercise : Let f be a measurable function defined on E, then its restriction to E1 is also
measurable where E1 is a measurable subset of E.
Solution : Let f1 = f/ E1 i.e., f1 is restriction of f to E1.
Let .
{x ∈ : f1(x) > } = { x ∈E1: f(x) > } [∵ = ]
Measurable Functions 43
= { x ∈E: f(x) > } ∩ E1is measurable on E and E1 is also measurable and
intersection of measurable sets is measurable. Hence f1 is measurable on E1.
2.33Exercise: .
.
Solution: Let f1 = f/ E1 and Let f2 = f/ E2
Let E =
Clearly E is measurable because . Suppose f is measurable on E then by
previous exercise f1 is measurable on E1 and f2 is measurable on E2.
Conversely, Let .
Therefore
{x ∈E: f(x) > } = { x ∈ : f(x) > }
= { x ∈E1: f(x) > }∪ { x ∈E2: f(x) > }
= { x ∈E1: f1(x) > }∪ { x ∈E2: f2(x) > }
because f1 is measurable on E1 and f2 is measurable on E2.
f is measurable on E = .
2.24 Theorem. If a function f is continuous almost everywhere in E, then f is measurable.
Proof. Since f is continuous almost everywhere in E, there exists a subset D of E with m*D = 0
such that f is continuous at every point of the set C = E-D.
To prove that f is measurable, let denote any given real number.
Consider the set {x ∈E | f(x) > } = B(say)
We have to show that B is measurable. If B ∩C = , ⊆ .
m*(B) ≤ m*(D) = 0.
m*(B) = 0.
B is measurable.
Now suppose that B ∩C ≠ . For this purpose, let x denote an arbitrary point in B ∩C. Then x ∈
∈ ⟹f(x) > and f is continuous at x. Hence there exists an open interval Ux
containing x such that f(y) > hold for every point y of E∩ Ux. Let U =∈∩ .
Since x ∈E ∩Ux ⊂B holds for every x ∈B ∩C, we have
B ∩C⊂E ∩Ux ⊂B. This implies B = (E ∩U) ∪(B ∩D). As an open subset of R, U is
measurable. Hence E ∩U is measurable. On the other hand, since m*(B∩D) ≤m*D = 0 ,B∩D is
also measurable. This implies that B is measurable. This completes the proof of the theorem.
2.25 Littlewood’s three principles of measurability
The following three principles concerning measure are due to Littlewood.
First Principle. Every measurable set is a finite union of intervals.
Second Principle. Every measurable function is almost a continuous function.
44 Measure and Integration Theory
Third Principle. If {fn} is a sequence of measurable function defined on a set E of finite measure and if
fn (x) →f(x) on E, then fn (x) converges almost uniformly on E.
First of all we consider third principle. We shall prove Egoroff’s theorem which is a slight modification
of third principle of Littlewood’s.
2.26 Theorem. Let E be a measurable set with finite measure and {fn} be a sequence of measurable
functions defined on a set E such that
fn (x) →f(x) for each x ∈ .
Then given > 0 and > , there corresponds a measurable subset A of E with m(A) < and an
integer N such that − < ∀ ∈ − ≥ .
Proof: Consider the sets = ∈ : − ≥
Now since fn and f are measurable.
So the sets are also measurable.
Now define = ∞
.
= : ∈ ≥
= : ∈ , − ≥ ≥
We observe that ⊆ .
On the contrary, we assume that for each ∈ ∀ .
Then for any fixed given k, we must have
= − ≥ ≥
But this leads to ↛f(x). a contradiction.
Hence for each x∈ there is some such that ∉ ⟹ ∞
= ∅
Now measure of E is finite, so by proposition of decreasing sequence, we have
∞
lm = ⋂ = ∅ =
→∞
lm = .
→∞
Hence given > , ∃ an integer N such that m( < ∀ ≥ .
In particular put k = N
<
: ∈ , − ≥ ≥ <
If we write A = , then m(A) <
E-A = : ∈ , − < ≥
Measurable Functions 45
In other words,
− < ≥ ∈ −
This completes the proof.
2.27 Definition: A Sequence {fn} of functions defined on a set E is said to converge almost everywhere
to f if lim f n ( x) f ( x) ∀ ∈ − ⊂ ,
n
= .
2.28 Theorem. Let E be a measurable set with finite measure and {fn} be a sequence of measurable
functions converging almost everywhere to a real valued function f defined on a set E. Then given
> 0 and > , there corresponds a measurable subset A of E with m(A) < and an integer N
such that − < ∀ ∈ − ≥ .
Proof: Let F be a set of points of E for which ↛f. Then m(F) = 0.
Since fn (x) →f(x) almost everywhere, then
fn (x) →f(x) ∀ ∈ − =
Now applying the last theorem for the set , we get a set A1 ⊆ E1 with m(A1) < and an integer N
such that − < ∀ ∈ − ≥ .
Now the required result follows if we take
= ∪ .
= ∪ = + = + = <
Also − = − ∪ = ∩ ∪
= ∩ ∩ = ∩ ∩
= − ∩ = ∩ = −
i.e., E-A = −
Hence we have found a set A ⊆ E with m(A) < and an integer N such that − < ∀ ∈
− ≥ .
2.29 Definition: A Sequence {fn} of functions is said to converge almost uniformly everywhere to a
measurable function f defined on a measurable set E if for each >0, ∃ A ⊆ E with
m(A) < such that and an integer N such that − .
2.30 Theorem.(Egoroff’s Theorem). Let {fn} be a sequence of measurable functions defined on a set E
of finite measure such that fn (x) →f(x) almost everywhere. Then to each > 0 there corresponds a
measurable subset A of E such that m (A) < such that fn(x) converges to f(x) uniformly on E-A.
Proof. Applying last theorem with = , = ⁄
We get a measurable subset A1⊆ E with m (A1) < ⁄ and positive integer N1 such that
− < ≥ ∈ = −
Again taking = , = ⁄
46 Measure and Integration Theory
We get another measurable subset A2⊆ E1 with m (A2) < ⁄ and positive integer N2 such that
− < ≥ ∈ = −
Continuing like that at kth stage, we get a measurable subset Ak⊆ Ek-1 with
m (Ak) < ⁄ and positive integer Nk such that
− < ≥ ∈ = −
Now we set A = ∞
Then we have
∞
m(A) ≤ ∞ ∞
< ⁄ = . = .
Also E-A = E- = − = ∵ − =
Let ∈ − , ∈ ∀ − < ∀ ≥ .
Choose k such that <
− < ∀ ∈ − ≥ = .
This completes the proof of the theorem.
Now we pass to the second principle of Littlewood. This is nothing but approximation of measurable
functions by continuous functions. In this connection we shall prove the following theorem known as
Lusin Theorem after the name of a Russian Mathematician Lusin, N.N.
2.31 Lusin Theorem: Let f be a measurable real valued function defined on closed interval [a,b],
then given > , ∃ ,
: ≠ < .
Proof: First we prove two lemmas.
Lemma 1. Let F be a closed subset of R, then a function g: F→ is continuous if sets : ≤
: ≥ are closed subsets of F for every rational a and b.
Proof: Let : ≤ : ≥ .
⇒ : > ∩ : < is open subset of F.
.i.e., : < < is open.
i.e., : ∈ , .
i.e., , .
Let O be any open set in R then O can be written as countable union of disjoint open intervals with
rational end points.
Measurable Functions 47
Let O = ∞
,
Then O = ∞ ∞
, = ,
Since , and countable union of open set is open.
⇒ ⇒ .
Lemma 2. Let f: [a, b] → be a measurable function, then given
> , ∃ = , − < .
Proof: Let .
For n∈ , take = : ≥
And ∗ = : ≤
Clearly each ∗ ∵
Then ∃ closed sets ⊂ ∗ ⊂ ∗
∗ ∗
− < − <
. .
∗ ∗
Let = ∞ ∞
− ∪ −
Clearly D is measurable.
∗ ∗
Therefore m(D) ≤ ∞ ∞
− + −
m(D) < ∞
+ ∞
. .
= + =
⇒ < .
Now E and D are measurable.
E-D is measurable.
Then for given > , ∃ ⊆ − − − <
Now E-F = D∪ − −
m(E-F) = m(D) + m − − < + =
Let h = f/F
To show that h is continuous on F.
For rational number ,
: ≤ = : ≤ ∩
= ∗ ∩ = ∗ − ∗ ∪ ∗ ∩
= ∗ − ∗ ∩ ∪ ∗ ∩
= ∅ ∪ ∗ ∩
= ∗ ∩
48 Measure and Integration Theory
∞ ∞
= ⋃ − ∪ ⋃ ∗ − ∗
∗ ∗
⇒ − ⊂
∵ ⊆ − ⇒ ∩ = ∅.
: ≤ = ∗ ∩
Since ∗ = , .
∗ ∩ .
: ≤ .
By lemma 1, h is continuous.
So f/F is continuous.
Lusin Theorem:(Proof):- We have
f:[a, b]→ is measurable function, then by lemma(2), for given > , ∃ ⊂
− < = .
Now using result “Every real valued continuous function defined on a closed subset of a real
number can be extended continuously to all real numbers.”
So h can be extended to continuous function h*: R→ .
Let g :[a, b]→ ,
∈ , = .
∈ : ≠ ⊆ −
∈ : ≠ ≤ − < .
“Convergence in Measure”
The notion of convergence in measure is introduced by F.Reisz and E.Fisher in 1906-07.
Sometimes it is also called approximate convergence.
2.32 Definition. A sequence < fn > of measurable functions is said to convergence in measure to f on a
set E, written as → ,
If given > 0, ∃ ∈ N such that for all n ≥ , we have
− ≥ < .
Or lm − ≥ =
→∞
This means that for all sufficiently large value of n, functions of the sequence < fn > differ from the
limit function f by a small quantity with the exception of the set of point whose measure is arbitrary
small (<.
2.33 Theorem: If sequence converges in measure to the function f, then it converges in
measure to every function g which is equivalent to the function.
Proof: For each > , we have
: − ≥ ⊂ : ≠ ∪ : − ≥
Measurable Functions 49
Since g is equivalent to f, then we have
: ≠ = .
: − ≥ ≤ : ≠ + : − ≥
≤ : − ≥ <
⇒ → .
Hence the result.
2.34 Theorem: If sequence converges in measure to the function f, then the limit function f is
unique a.e.
Proof: Let g be another function such that → .
Since − ≤ − + −
Now we observe that for each > ,
: − ≥ ⊂ : − ≥ ∪ : − ≥
Since by proper choice of , the measure of both the sets on the right can be made arbitrary small, we
have
: − ≥ =0
f = g almost everywhere. Hence the proof.
2.35 Theorem: Let be a sequence of measurable functions which converges to f a.e. on
X. Then → .
Proof: For each n∈ > ,
= ∈ : − ≥
Let > , ∃ ⊂
With m(A) < and the number N such that
− < ∀ ∈ − ≥
Then it follows that ⊂ ∀ ≥
< < ∀ ≥
lm =
→∞
Hence → .
2.36 Remark: The converse of the above theorem need not be true i.e, convergence in measure is
more general than a.e. infact there are sequence of measurable functions that converges in
measure but fails to converge at any point.
To affect we consider the following example
: , →
+
= , ∈ ,
,
50 Measure and Integration Theory
Let n = + ≤ ≤ .
Let > . Choose an m ∈ <
Then : − ≥ = : ≥
∵ = + < +
= < [ ] (*)
< . , <
≤ < ∀ ≥
converges in measure to zero for x∈ ,
. . , → ,
fn (x) has value 1 for arbitrary large value of n and so it does not converge to zero a.e. because
on taking n very large, we get 2t large and hence number of subintervals of type (*) increase and
possibility of = .
2.37 Theorem (F. Riesz). “Let < fn > be a sequence of measurable functions which converges in
measure to f. Then there is a subsequence < > of < fn > which converges to f almost
everywhere.”
Proof. Let → .
Let us consider two sequences
∞
→ → ∑ = < .
We now choose a strictly increasing sequence { } of positive integer as follows
Let be a positive integer such that
: 1 − ≥ <
Such a number exists since in view → for a given = >
= > , ∃
: − ≥ < ∀ ≥
In particular for n = .
Similarly, Let be a positive number such that ≥
: 2 − ≥ < .
Continuing in this process, we get the positive number ≥
: − ≥ <
Measurable Functions 51
Now set Ek = ∞
: − ≥ , ∈ .
And = ∞
Then it is clear that is decreasing sequence of measurable sets.
Therefore = lm
→∞
But = ∞
: − ≥
≤ ∞
: − ≥
< ∞
→ →
=
1
Hence m(E) = 0.
Thus it remains to be verified that the sequence < > converges to f on X-E.
So let ∉ . Then ∉ for some positive integer m.
i.e., ∉ : − ≥ , ≥
⇒ − < , ≥
But → as k→
Hence lm = .
→∞
Since ∈ − was arbitrary, it follows that
lm = for each x ∈ − and so . .
→∞
This completes the proof.
SECTION – III
THE LEBESGUE INTEGRAL
Lebesgue integration is an alternative way of defining the integral in terms of measure theory that is
used to inte grate a much broader class of functions than the Riemann integral or even the Riemann-
Stieltjes integral. The idea behind the Lebesgue integral is that instead of approximating the total area by
dividing it into vertical strips, one approximates the total area by dividing it into horizontal strips.
3.1 The shortcomings of the Riemann integral suggested the further investigations in the theory of
integration. We give a resume of the Riemann Integral first.
Let f be a bounded real- valued function on the interval [a, b] and let
= ξ < ξ < ⋯ < ξ =
Be a partition of [a, b]. Then for each partition we define the sums
= ξ − ξ
and s = ξ − ξ m
where
= up , m =
ξi1 xξi ξi1 xξi
We then define the upper Riemann integral of f by
b
a d = inf S
With the infimum taken over all possible subdivisions of [a, b].
Similarly, we define the lower integral
b
a d = sup s.
The upper integral is always at least as large as the lower integral, and if the two are equal we say that f
is Riemann integrable and call this common value the Riemann integral of f. We shall denote it by
b
a
To distinguish it from the Lebesgue integral, which we shall consider later.
By a step function we mean a function ψ which has the form
ψ = , ξ < < ξ
for some subdivision of [a, b] and some set of constants ci .
The Lebesgue Integral 53
The integral of ψ is defined by
b
∫ ψ d = ∑ ξ − ξ .
a
With this in mind we see that
b b
a d = inf a ψ d
for all step function ψ ≥ .
Similarly,
b b
a d = sup a d
for all step functions ≤ .
3.2. Example: If
s tol
f(x) =
o tol
b b
then a d = − and a d = .
Thus we see that f(x) is not integrable in the Riemann sense.
3.3. The Lebesgue Integral of a bounded function over a set of finite measure
The example we have cited just now shows some of shortcomings of the Riemann integral. In particular,
we would like a function which is 1 in measurable set and zero elsewhere to be integrable and have its
integral the measure of the set.
The function χ defined by
χ =
o ∉
is called the characteristic function on E. A linear combination
= χ
is called a simple function if the sets Ei are measurable. This representation for is not unique.
However, we note that a function is simple if and only if it is measurable and assume only a finite
number of values. If is simple function and [a1 , a2 ,…, an ] the set of non- zero values of , then
= χA ,
i
where = = . This representation for is called the canonical representation and
it is characterized by the fact that the Ai are disjoint and the ai distinct and non- zero.
If vanishes outside a set of finite measure, we define the integral by
∫ d = ∑ m
54 Measure and Integration Theory
when has the canonical representation = χA . we sometimes abbreviate the expression for
i
this integral to . If E is any measurable set, we define ∅ = . χ
It is often convenient to use representations which are not canonical, and the following lemma is useful.
3.4. Lemma. If E1, E2,…,En are disjoint measurable subset of E then every linear combination
= χ
i
With real coefficients c1, c2,…, cn is a simple function and
= m .
Proof. It is clear that is a simple function. Let a1, a2,…, an denote the non- zero real number in .
For each j = 1, 2, …, n . Let
= ⋃
i aj
Then we have = = =
and the canonical representation
= ∑ χA
j
Consequently, we obtain
= m
= m i aj
= i aj m ( Since Ei are disjoint, additivity of measures applies )
∑ m
This completes the proof of the theorem.
3.5. Theorem. Let and ψ be simple functions which vanish outside a set of finite measure. Then
+ ψ = + ψ and, if ≥ ψ a.e., then ≥ ψ
Proof. Let {Ai} and {Bi} be the sets which occur in the canonical representations of and ψ. Let A0
and B0 be the sets where and ψ are zero. Then the sets Ek obtained by taking all the intersection ∩
form a finite disjoint collection of measurable sets, and we write
N
= ∑ χ
k
The Lebesgue Integral 55
N
ψ = ∑ χ
k
and so
N N
+ ψ = ∑ χ + ∑ χ
k k
N N
= ∑ χ + ∑ χ
k k
N
= ∑ + χ
k
Therefore
N
+ ψ = ∑ + m
N N
= ∑ m + ∑ m
= + ψ .
To prove the second statement, we note that
∫− ∫ψ = ∫ − ψ ≥ ,
Since the integral of a simple function which is greater than or equal to zero almost everywhere is
non- negative by the definition of the integral.
3.6. Remark. We know that for any simple function we have
N
= ∑ χ
i
Suppose that this representation is neither canonical nor the sets Ei’s are disjoint. Then using the fact
that characteristics functions are always simple function we observe that
∫ = ∫ χ + ∫ χ + … + ∫ χ
1 2 n
= χ E + χ E + ⋯ + χ E
1 2 n
= m + m + ⋯ + m
56 Measure and Integration Theory
N
= ∑ m
Hence for any representation of , we have
N
∫ = ∑ m
Let f be a bounded real valued function and E be a measurable set of finite measure. By analogy with the
Riemann integral we consider for simple functions and ψ the numbers
∫ ψ
ψf
and
sup ∫
ϕf
and ask when these two numbers are equal. The answer is given by the following proposition .
3.7. Theorem. Let f be defined and bounded on a measurable set E with mE finite. In order that
∫ ψ d = sup ∫ ψd
fψ fψ
For all simple functions and ψ, it is necessary and sufficient that f be measurable.
Proof. Let f be bounded by M and suppose that f is measurable. Then the sets
KM KM
= | ≥ > , − ≤ ≤ ,
Are measurable, disjoint and have union E. Thus
∑ m = m
The simple function defined by
ψ = ∑ χ
k
and
= ∑ −χ
k
satisfy
≤ ≤ ψ
The Lebesgue Integral 57
M
Thus ψd ≤ ψ d = m
M
and sup d ≥ d = − m
M M
hence ≤ ψd − sup d ≤ m = m .
Since n is arbitrary we have
∫ ψd − sup ∫ d = ,
and the condition is sufficient.
Suppose now that ψd = sup d .
ψf ϕf
Then given n there are simple functions d ψ such that
≤ ≤ ψ
And (1) ψ d − d <
Then the functions ψ∗ = ψ
And ∗ = sup
Are measurable and ∗ ≤ ≤ ψ∗ .
Now the set
Δ = ∗ < ψ∗
is the union of the sets
Δv = | ∗ < ψ∗ − .
But each Δv is contained in the set | < ψ − , and this latter set by (1) has measure less
v
v
than . Since n is arbitrary, mΔv = and so mΔ = . Thus ∗ = ψ∗ except on a set of measure zero,
and ∗ = except on a set of measure zero. Thus f is measurable and the condition is also necessary.
3.8. Definition. If f is a bounded measurable function defined on a measurable set E with mE finite, we
define the Lebesgue integral of f over E by
∫ d = ∫ ψ
for all simple functions ψ ≥ .
By previous theorem, this may also be defined as
58 Measure and Integration Theory
∫ d = sup ∫
for all simple functions ≤ .
b
We sometime write the integral as . If E =[a,b] we write a instead of a,b .
Definition and existence of the Lebesgue integral for bounded functions
3.9. Definition. Let F be a bounded function on E and let Ek be a subset of E. Then we define M[f, Ek]
and m[f, Ek] as
, = l. u.
xεk
m, = . l.
xεk
3.10. Definition. By a measurable partition of E we mean a finite collection P = {E1, E2, …, En} of
measurable subsets of E such that
⋃ =
And such that m ∩ = , = ,, … , , ≠
The sets E1 , E2 ,…,En are called the components of P.
If P and Q are measurable partitions, then Q is called a refinement of P if every component of Q is
wholly contained in some component of P.
Thus a measurable partition P is a finite collection of subsets whose union is all of E and whose
intersections with one another have measure zero.
3.11. Definition. Let f be a bounded function on E and let P={ E1 , E2 ,…,En} be any measurable
partition E. we define the upper sum U[f, P] as
U; = ∑ ; . m
Similarly, we define the lower sum L[f; P] as
L; = ∑ m; . m
As in the case of Riemann integral, we can see that every upper sum for f is greater than or equal to
every lower sum for f.
We then define the Lebesgue upper and lower integral of a bounded function f on E by
U; d sup L;
P P
The Lebesgue Integral 59
Respectively taken over all measurable position of E. We denote them respectively by
∫ d ∫
̅
3.12. Definition. We say that a bounded function f on E is Lebesgue integrable on E if
∫ d ∫
̅
Also we know that if ψ is a simple function, then
∫ ψ = ∑ m
Keeping this in mind, we see that
∫ = ∫ ψd
For all simple functions ψ ≥ . Similarly
∫ = sup ∫ d
̅
̅
For all simple functions ≤ .
Now we use the theorem :
“ Let f be defined and bounded on a measurable set E with mE finite. In order that
∫ ψd = sup ∫ d
fψ fϕ
for all simple functions and ψ, it is necessary and sufficient that f is measurable.”
And our definition of Lebesgue integration takes the form :
“ If f is a bounded measurable function defined on a measurable set E with mE finite , we define the
(Lebesgue) integral of f over E by
∫ d = ∫ ψd
for all simple functions ψ ≥ .”
The following theorem shows that the Lebesgue integral is in fact a generalization of the Riemann
integral.
60 Measure and Integration Theory
3.13. Theorem. Let f be a bounded function defined on [a,b]. If f is Riemann integrable on [a, b], then it
is measurable and
b b
∫ d = ∫
a a
Proof . Since f is a bounded function defined on [a, b] and is Riemann integrable, therefore,
̅
b b
∫ d = ∫ d
ϕf
a a
and
b b
∫ d = sup ∫ ψd
ψf
a a
for all step functions and ψ and then
̅
b b
∫ d = ∫ d
a a
b b
⇒ a d = sup a ψd (i)
ϕf ψf
Since every step function is a simple function, we have
b b b b
∫ d = sup ∫ ψd ≤ ∫ d = ∫ d
ψf ϕf
a a a a
Then (i) implies that
b b
sup ∫ ψd = ∫ d
ψf ϕf
a a
and this implies that f is measurable also.
3.14. Comparison of Lebesgue and Riemann integration
(1) The most obvious difference is that in Lebesgue’s definition we divide up the interval into subsets
while in the case of Riemann we divide it into subintervals.
(2) In both Riemann’s and Lebesgue’s definitions we have upper and lower sums which tend to limits.
In Riemann case the two integrals are not necessarily the same and the function is integrable only if
they are same. In the Lebesgue case the two integrals are necessarily the same, their equality being
consequence of the assumption that the function is measurable.
The Lebesgue Integral 61
(3) Lebesgue’s definition is more general than Riemann. We know that if function is the R- integrable
then it is Lebesgue integrable also, but the converse need not be true. For example the characteristic
function of the set of irrational points have Lebesgue integral but is not R- integrable.
Let χ be the characteristic function of the irrational numbers in [0,1]. Let E 1 be the set of irrational
number in [0,1], and let E2 be the set of rational number in [0,1]. Then P = [E1, E2] is a measurable
partition of (0,1]. Moreover, χ is identically 1 on E1 and χ is identically 0 on E2 . Hence M[χ, E1] =
m[χ, E2] = 1 , while M[χ, E1] = m[χ, E2] = 0. Hence U[χ, P] = 1.m E1 + 0.m E2 = 1. Similarly L[χ, P]
= 1.m E1 + 0.M E2 = 1. Therefore, U[χ, P] = L[χ, P] .
For Riemann integration
M[χ,J] = 1 , m[χ,J] = 0
for any interval J ⊂ ,
∴ Uχ, J = , Lχ, J = .
∴ The function is not Riemann- integrable.
3.15. Theorem. If f and g are bounded measurable functions defined on a set E of finite measure, then
(i) =
(ii) + = +
(iii) If ≤ . e. , te ≤
(iv) If = . e. , te =
(v) If ≤ ≤ , te m ≤ ≤ m.
(vi) If A and B are disjoint measurable set of finite measure, then A∪B = A + B
Proof. We know that if ψ is a simple function then so is a ψ.
Hence = ψ = ψ =
ψf ψf
Which proves (i).
To prove (ii) let ε denote any positive real number. These are simple functions ≤ , ψ ≥ , ξ ≤
d η ≥ satisfying
∫ d > ∫ − ε , ∫ ψd < ∫ + ε ,
∫ ξd > ∫ − ε , ∫ η < ∫ + ε,
Since + ξ ≤ + ≤ ψ + η , we have
∫ + ≥ ∫ + ξ = ∫ + ∫ ξ > ∫ + ∫ − ε
∫ + ≤ ∫ψ + η = ∫ ψ + ∫ η < ∫ + ∫ + ε
62 Measure and Integration Theory
Since these hold for every ε > , we have
∫ + = ∫ + ∫
To prove (iii) it suffices to establish
∫− ≥
For every simple function ψ ≥ − , we have ψ ≥ almost everywhere in E. This means that ψ ≥
Hence we obtain
∫ − = ∫ ψ ≥
ψgf
Which establishes (iii).
Similarly we can show that
∫ − = sup ∫ ψ ≤
ψgf
Therefore, from (1) and (2) the result (iv) follows.
To prove (v) we are given that
≤ ≤
Applying (iv) we are given that
∫ d ≤ ∫ d = ∫ d = m
That is, ≤ m
Similarly we can prove that ≥ m.
Now we prove (vi).
We know that χA∪B = χA + χB
Therefore , A∪B = A∪B χA∪B = A∪B χA + χB
= ∫ χA + ∫ χB
A∪B A∪B
= ∫+ ∫
A B
Which proves the theorem.
The Lebesgue Integral 63
3.16. Corollary. If f and g are bounded measurable function then
If ≥ on E then ≥ and
If ≤ on E then ≤ .
Proof : Let be a simple function such that ≥
Since ≥ on E ⇒ ≥ on E
⇒ ≥ ⇒ ≥ i.e. ≥
ψf
Similarly, Let be a simple function such that ≤ . Since ≥ on E
⇒ ≤ on E
⇒ ≤ ⇒ up ≤ i.e. ≤
ϕf
3.17. Corollary. If m(E) = 0 , then =
Or
Integrals over set of measure zero are zero.
Proof : Since f is bounded on E so there exist constant A and B such that
≤ ≤
⇒ . m ≤ ∫ d ≤ . m ∀ ∈
Since m(E) = 0 ⇒ =
3.18. Corollary. If f(x) = k a.e. on E then = . m . In particular if f = 0 a.e. on E then =
Proof : Since f(x) = k a.e on E then =
3.19. Corollary. If f = g a.e then = but converse is not true.
Proof : consider the functions
∶ −, ⟶ and ∶ , ⟶
≤
as = and g(x) =1 ∀
>
Clearly f and g are bounded and measurable functions.
⇒ f and g are lebesgue integrable on [-1,1]
∫ d = ∫ d + ∫ d
= d + . d =
64 Measure and Integration Theory
∫ = ∫ d = . m−, = . =
Therefore =
But ≠ a.e on [-1,1]
{ ∵ m ∈ −, ; ≠ = ≠
Therefore ≠ a.e on E
f and g are not equal even at a single point of [-1,1] as these are defined.
3.20. Corollary. If f = 0 a.e on E then = but converse is not true.
≥
Proof : Consider the function f : [-1,1] → as =
− <
∫ d = ∫ d + ∫ d
= -1 + 1 = 0
Clearly ≠ a.e as m ∈ −, ; ≠ = m−, = ≠
So converse is not true.
3.21.Corollary. If = and ≥ on E then f = 0 a.e.
Proof : Suppose E has a subset A where f(x) > 0,
i.e. = ∞
x ∈ ; >
Let = ∈ ; >
If possible, suppose there is a positive integer N such that m( > .
Then ≥ ≥ m( >
1 N N
Which contradicts the fact that =
Thus, m( = for all n≥ .
This proves the corollary.
3.22. Corollary. Let f be a bounded measurable function on a set of finite measure E. Then
≤
Proof : The function is measurable and bounded
Now - ≤ ≤ on E
The Lebesgue Integral 65
By the linearity and monotonicity of integration,
- ≤ ≤
⇒ ≤
3.23. The Monotone Convergence Theorem
Let {fn} be an increasing sequence of non-negative measurable functions on E. If {fn} → pointwise a.e
on E, then lm =
→∞
Proof : Since {fn} is an increasing sequence
So fn ≤ a.e ∀
⇒ lm ≤ … (1)
Now by Fatou’s Lemma ≤ lm … (2)
From (1) and (2), we have
lm ∫ = lm ∫
Hence the result .
Case II If f is a bounded function on E, then theorem is trivially true. Since in this case
∈
≤ ∀ ∈ for some number M and thus ∈ > , one can choose a δ = ( ) > for which m(A)
M
< δ, then A ≤ A = . m < ∈.
3.24. Remark : The technique used in above theorem helps us to evaluate the lebesgue integral of non-
negative bounded and unbounded functions.
3.25. Example : Evaluate the Lebesgue integral of the function f : [0,1] →
⁄ < ≤
f(x) = ⁄
=
Clearly f is unbounded, non-negative function defined on [0,1]. Now define a sequence of functions {fn}
on [0,1] as
≤
fn(x) =
<
≥
3
i.e. fn(x) = {
<
3
Clearly {fn} is increasing sequence of non-negative measurable functions such that fn→ . So by
monotone convergence theorem
66 Measure and Integration Theory
∫ d = lm ∫ d
→∞
⁄
= lm 3 d + ⁄ d
→∞ 3
⁄
3
= lm ∫ d + ∫ d
→∞
⁄
3
⁄
3 ⁄
= lm + ∫ d
→∞
⁄
3
= lm .
+ −
→∞
=+ =
3.26. Theorem(Lebesgue Bounded Convergence Theorem). Let < fn > be a sequence of measurable
functions defined on a set E of finite measure and suppose that < fn > is uniformly bounded, that is ,
there exist a real number M such that ≤ for all ε and for all ε . If lm = for
→∞
each x in E, then
∫ = lm ∫ .
→∞
Proof. We shall apply Egoroff’s theorem to prove this theorem. Accordingly for a given ε > , there is
ϵ
an N and a measurable set ⊂ such that m < and for ≥ and ε we have
4M
− <
m
∫ − ∫ = ∫ − ≤ ∫ −
= ∫ − + ∫ −
0 c0
< . m +
m
< + = .
Hence
∫ → ∫
The Lebesgue Integral 67
3.27. Remark : Bounded Convergence Theorem need not be true in Riemann integral .
3.28. Example : Let {ri} be a sequence of all rational numbers in [0,1].
Define = ∶ = ,, … , , ∈
∈
and for each ∈ , consider the function = = , , … ,
∉
clearly each is bounded, also is discontinuous at n-points in [0,1] namely points of
i.e., , , … , .
At =
lm ≠ ≠ lm
x→r
1 x→r1
Hence Riemann integrable on [0,1]
∵ uto s em tele, t s otous eept t te ume o dsotut
Now we have proved that
lm ∫ d ≠ ∫ lm d
→∞ →∞
⇒ ∫ d = ∫ d = ∫ d
Sn ∪Scn
∵ ∪ = ,
= ∫ d + ∫ d
Sn Scn
= ∫ d + ∫ d = . m =
Sn Scn
∵ sequee o tols m =
⇒ lm ∫ d =
→∞
Clearly {fn} is convergent to f when f is defined as
s tol ,
=
s tol ,
and f is not Riemann – integrable on [0,1]
⇒ ∫ d does ot ests .
68 Measure and Integration Theory
lm ∫ d ≠ ∫ lm d
→∞ →∞
So bounded convergence theorem does not hold in Riemann integral .
The integral of a non-negative function
3.29. Definition. If f is a non-negative measurable function defined on a measurable set E, we define
∫ = sup ∫ ,
hf
Where h is a bounded measurable function such that m ≠ is finite.
3.30. Theorem. If f and g are non-negative measurable functions, then
(i) = >
(ii) + = + and
(iii) ≤ . e. , te
∫ ≤ ∫
Proof. The proof of (i) and (iii) follow directly from the theorem concerning properties of the integrals
of bdd functions.
We prove (ii) in detail.
If ≤ and ≤ , we have + ≤ + , and so
∫ + ≤ ∫ +
i.e. + ≤ + .
Taking suprema, we have
(iv) + ≤ +
On the other hand, let ℓ be a bounded measurable function which vanishes outside a set finite measure
and which is not greater than + . Then we define the functions h and k by setting
= m, ℓ
and = ℓ −
we have ≤ ,
≤
while h and k are bounded by the bound ℓ and vanish where ℓ vanishes. Hence
The Lebesgue Integral 69
∫ℓ = ∫+∫ ≤ ∫ +∫
And so taking supremum, we have
sup ∫ ℓ ≤ ∫ + ∫
ℓfg
That is,
(v) + ≥ +
From (iv) and (v), we have
∫ + = ∫ + ∫
3.31. Fatou’s lemma. If < fn >is a sequence of non-negative measurable functions and fn(x) → f(x)
almost everywhere on a set E, then
∫ ≤ lm ∫
Proof. Let h be a bounded measurable function which is not greater than f and which vanishes outside a
set E’ of finite measure. Define a function hn by setting
hn(x) = min{h(x), fn(x)}
Then hn is bounded but bounds for h and vanishes outside E’ . Now hn(x) →h(x) for each x in E’ .
Therefore by “Bounded Convergence theorem” we have
∫ = ∫ = lm ∫ ≤ lm ∫
Taking the supremum over h, we get
∫ ≤ lm ∫
3.32. The inequality in Fatou’s lemma may be strict
Consider a sequence {fn} defined on R as
∈ , +
fn(x) =
otewse
Clearly sequence {fn} is sequence of non – negative measurable funcyions defined on R and
lm = where f = 0 ⇒ R =
→∞
70 Measure and Integration Theory
Also R = = +
1 2 1 2
= + = m =
1
⇒ lm R = and we know that 0 < 1
So R < lm R
3.33. Fatou’s lemma need not good unless the function fn is non – negative
− ≤ ≤
Let us consider the function fn(x) =
otewse
Hence lm = = a.e ⇒ d =
→∞
Also d = d + d
1 2
⁄
=⁄ − d + = −
Thus lm d = −
⇒ ∫ d ≰ lm ∫ d
3.34. Theorem( Lebesgue Monotone Convergence theorem). Let < fn > be an increasing sequence of
non negative measurable functions and let f = lim fn . Then
∫ = lm ∫
Proof. By Fatou’s Lemma we have
∫ ≤ lm ∫
But for each n we have fn ≤ f , son ≤ . But this implies
∫ ≤ ∫
Hence
∫ = ∫
3.35. Definition. A non-negative measurable functions f is called integrable over the measurable over
the measurable set E if
∫ <
The Lebesgue Integral 71
3.36. Theorem. Let f and g be two non-negative measurable functions. If f is integrable over E and
g(x)<f(x) on E, then g is also integrable on E, and
∫ − = ∫ − ∫
Proof. Since = − +
and the left hand side is finite, the term on the right must also be finite and so g is integrable.
3.37. Theorem. Let f be a non-negative function which is integrable over a set E. The given > there
is a > such that for every set ⊂ with < we have
<
Proof. If ≤ , then ≤ =
Set < Then < . = .
Set fn(x) = f(x) if f(x) ≤ n and fn(x) = n otherwise. Then each fn is bounded and fn converges to f at each
point. By the monotone convergence theorem there is an N such that > −
− < .
Choose < . If mA < , we have
∫ = ∫ − + ∫
< ∫ − +
∫ ≤ ∫ =
< + = .
3.38. The General Lebesgue Integral
We have already defined the positive part and negative part of a function as
= ,
= −,
Also it was shown that
= −
= +
With these notions in mind, we make the following definition.
72 Measure and Integration Theory
3.39. Definition. A measurable function f is said to be integrable over E if and are both integrable
over E. In this case we define
∫ = ∫ − ∫
3.40. Theorem. Let f and g be integrable over E. Then
(i) The function f+g is integrable over E and
∫ + = ∫ + ∫
(ii) If ≤ . ., then
∫ ≤ ∫
(iii) If A and B are disjoint measurable sets contained in E, then
∫ = ∫ + ∫
∪
Proof. By definition, the function , , , are all integrable. If = + , then = −, +
− and hence = + − + . Since + and + are integrable therefore
their difference is also integrable. Thus h is integrable.
We then have
∫ = ∫ + − +
= ∫ + − ∫ +
= ∫ + ∫ − ∫ − ∫
= ∫ − ∫ + ∫ − ∫
That is, + = +
Proof of (ii) follows from part (i) and the fact that the integral of a non-negative integrable function is
non-negative.
For (iii) we have ∪ = ∪
The Lebesgue Integral 73
= ∫ + ∫ = ∫ + ∫
∗It should be noted that f+g is not defined at points where f = and g = − and where f = − and
g = . However, the set of such points must have measure zero, since f and g are integrable. Hence the
integrability and the value of + is independent of the choice of values in these ambiguous cases.
3.41. Theorem. Let f be a measurable function over E. Then f is integrable over E iff is integrable
over E. Moreover, if f is integrable, then
∫ = ∫
Proof. If f is integrable then both and are integrable. But = + . Hence integrability of
and implies the integrability of .
Moreover, if f is integrable, then since ≤ = , the property which states that if ≤
a.e. , then ≤ implies that
∫ ≤ ∫
On the other hand since − ≤ , we have
− ≤ .
From (i) and (ii)
Conversely, suppose f is measurable and suppose is integrable. Since
≤ ≤
It follows that is integrable. Similarly is also integrable and hence f is integrable.
3.42. Lemma. Let f be integrable . Then given > there exist > such that < whenever
A is measurable function f we have = − . So by that we have proved already, given > , there
exist > such that
∫ < .
When mA< . Similarly there exists > such that
∫ < ,
When mA< . Thus if mA < = min ( , , we have
≤ = + < + =
This completes the proof.
74 Measure and Integration Theory
3.43. Theorem (Lebesgue Dominated Convergence Theorem) Let a sequence < fn > , of
measurable functions be dominated by an integrable function g, that is
≤
Holds for every and every and let < fn > converges pointwise to a function f, that is , f(x) =
for almost all x in E. Then
→∞
∫ = ∫
→∞
Proof. Since ≤ for every and f(x) = , we have ≤ . Hence fn and f are
→∞
integrable. The function g – fn is non-negative, therefore by Fatou’s Lemma we have
∫ − ∫ = ∫ − ≤ ∫ −
= ∫ − ∫
Whence ≥ lm
Similarly considering g + fn we get
∫ ≤ lm ∫
Consequently, we have = lm
SECTION – IV
DIFFERENTIATION AND INTEGRATION
The “fundamental theorem of the integral calculus” is that differentiation and integration are inverse
processes. This general principle may be interpreted in two different ways.
If f(x) is integrable, the function
F(x) =
is called the indefinite integral of f(x); and the principle asserts that(i) ́ (x) = f(x) (ii)
On the other hand, if F(x) is a given function, and f(x) is defined by (ii), the principle asserts that
x
a tdt = F(x) − F(a) (iii)
The main object of this chapter is to consider in what sense these theorems are true.
From the theory of Riemann integration (ii) follows from (i) if x is a point of continuity of f. For we can
choose h0 so small that |f(t) −f(x)| < for |t−x| ≤ h0; and the
ℎ xh
− = t − dt ≤ ( |h| < h0) by the mean-value theorem.
ℎ ℎ x
This proves (ii).
We shall show that more generally this relation holds almost everywhere. Thus differentiation is the
inverse of Lebesgue integration.
The problem of deducing (iii) from (ii) is more difficult and even using Lebesgue integral it is true only
for a certain class of functions. We require in the first place that ́ (x)should exist at any rate almost
everywhere and as we shall see this is not necessarily so. Secondly, if ́ (x) exists we require that it
should be integrable.
Differentiation of Monotone Functions
4.1. Definition. Let C be a collection of intervals. Then we say that C covers a set E in the sense of
Vitali, if for each > 0 and x in E there is an interval I ∈ C such that x∈I and l(I) < .
Now we prove the following lemma which will be utilized in proving a result concerning the
differentiation of monotone functions.
4.2. Lemma. (Vitali). Let E be a set of finite outer measure and C a collection of intervals which cover
E in the sense of Vitali. Then given > 0 there is a finite disjoint collection {I1,…, In} of intervals in C
such that
m*[E− ∪
] < .
76 Measure and Integration Theory
Proof. It suffices to prove the lemma in the case that each interval in C is closed, for otherwise we
replace each interval by its closure and observe that the set of endpoints of I1, I2,…, IN has measure zero.
Let O be an open set of finite measure containing E. Since C is a Vitali covering of E, we may suppose
without loss of generality that each I of C is contained in O. We choose a sequence <In > of disjoint
intervals of C by induction as follows :
Let I1 be any interval in C and suppose I1,…, In have already been chosen. Let kn be the supremum of
the lengths of the intervals of C which do not meet any of the intervals I1,…,In .
Since each I is contained in O, we have kn ≤ m O < . Unless, E ⊂ ∪ I we can find In+1 in C
with l(In+1) > kn and In+1 is disjoint from I1, I2,…, In. Thus we have a sequence < In > of disjoint
intervals of C, and since U In ⊂O, we have In) ≤ m < .
Hence we can find an integer N such that ∞
<
Let R=E− ∪
.
It remains to prove that m*R < .
Let x be an arbitrary point of R. Since ∪
is a closed set not containing x, we can find an
interval I in C which contains x and whose length is so small that I does not meet any of the intervals
I1, I2,…, IN . If now I ∩ Ii = ∅ for i ≤ N, we must have l(I) ≤ kN < 2l (IN+1). Since lim l(In) = 0 , the
interval I must meet at least one of the intervals In. Let n be the smallest integer such that I meets In.
We have n > N, and l(I) ≤ kN ≤ 2l (IN+1). Since x is in I, and I has a point in common with In, it follows
that the distance from x to the midpoint of In is at most l(I) + l (IN) ≤ l (IN+1).
Let Jm denote the interval which has the same midpoint as Im and five times the length of Im. Then
we have x ∈ Jm. This proves R ⊂∪∞ J
Hence m* R≤ ∞ ∞
= < .
The Four Derivatives of a Function
Whether the differential coefficients
fℎf
́ (x) = lm
ℎ→ ℎ
exist or not, the four expressions
fℎf
D+f(x) = lm
ℎ→ ℎ
ffℎ
D f(x) = lm
ℎ→ ℎ
fℎf
D f(x) = lm ℎ→
ℎ
ffℎ
D f(x) = lm ℎ→
ℎ
Differentiation and Integration 77
always exist. These derivatives are known as Dini Derivatives of the function f.
D+ f(x) and D+ f(x) are called upper and lower derivatives on the right and D f(x) and D f(x) are called
upper and lower derivatives on the left. Clearly we have D+ f(x) ≥ D+ f(x) and D f(x) ≥ D f(x).
If D+ f(x) = D+ f(x), the function f is said to have a right hand derivative and if D f(x) = D f(x), the
function is said to have a left hand derivative.
If D+ f(x) = D+ f(x) = D f(x) = D f(x)≠ ∓ we say that f is differentiable at x and define f’(x) to
be the common value of the derivatives at x.
4.3.Theorem. Every non-decreasing function f defined on the interval [a, b] is differentiable almost
everywhere in [a, b]. The derivative f’ is measurable and
b
a d = f(b) − f(a)
Proof. We shall show first that the points x of the open interval (a, b) at which not all of the four Dini-
derivatives of f are equal form a subset of measure zero. It suffices to show that the following four
subsets of (a, b) are of measure zero:
A = {x ∈ (a, b) | D- f(x) < D+ f(x) },
B = {x ∈ (a, b) | D+ f(x) < D- f(x) },
C = {x ∈ (a, b) | D- f(x) < D- f(x) },
D = {x ∈ (a, b) | D+ f(x) < D+ f(x) }. To prove m* A = 0 , consider the subsets
Au,v = {x ∈ (a, b) | D- f(x) < u < v < D+ f(x) }
of A for all rational numbers u and v satisfying u < v. Since A is the union of this countable family
{Au,v}, it is sufficient to prove m* (Au,v) = 0 for all pairs u, v with u < v.
For this purpose, denote = m* (Au,v) and let be any positive real number. Choose an open set
U ⊃ Au,v with m* U < + . Set x be any point of Au,v . Since D- f(x) < u, there are arbitrary small
closed intervals of the form [x-h, x] contained in U such that
f(x) - f(x-h) < uh.
Do this for all x ∈ Au, v and obtain a Vitali cover C of Au,v. Then by Vitali covering theorem there is a
finite subcollection {J1, J2,…, Jn}of disjoint intervals in C such that
m*(Au,v - ∪ J ) <
Summing over these n intervals, we obtain
− <
< u m*U
< u( + )
Suppose that the interiors of the intervals J1, J2,…, Jn cover a subset F of Au,v. Now since
D+ f(y) > v, there are arbitrarily small closed intervals of the form [y, y+k] contained in some of the
intervals Ji (i = 1, 2,…, n) such that
78 Measure and Integration Theory
f(y+k) − f(y) > vk
Do this for all y ∈ F and obtain a Vitali cover D of F. Then again by Vitali covering lemma we can
select a finite subcollection [K1, K2, …, Km] of disjoint intervals in D such that
m* [F− ∪
Ki ] <
Since m*F > − , it follows that the measure of the subset H of F which is covered by the intervals is
greater than − . Summing over these intervals and keeping in mind that each Ki is contained in a Jn,
we have
− − ≥ ]
+ −
> v
> −
So that
− < −
Since this is true for every > 0 , we must have v < u . Since u < v, this implies that = 0 . Hence
m*A = 0. Similarly, we can prove that m*B = 0, m*C = 0 and m*D = 0. This shows that
fℎf
g(x) = lm
ℎ→ ℎ
is defined almost everywhere and that f is differentiable whenever g is finite.
If we put
gn(x) = n[f (x+ ) - f(x)] for x ∈ [a,b] ,
where we re-define f(x) = f(b) for x ≥ b. Then gn(x) → g(x) for almost all x and so g is measurable since
every gn is measurable. Since f is non-decreasing, we have gn ≥ 0. Hence, by
Fatou’s Lemma
b
≤ = n a ( + ) − d
1
b b
= n 1
n
d − a d
an
1 1
b b a b
= n a + b n d − a n d − a d
b
= n a ( + ) − d
≤ f(b)-f(a)
(Use of f(x) = f(b) for x ≥ b for first interval and f non-decreasing in the 2nd integral).
Differentiation and Integration 79
This shows that g is integrable and hence finite almost everywhere. Thus f is differentiable almost
everywhere and g(x) = ́ (x) almost everywhere. This proves the theorem.
Functions of Bounded Variation
Let f be a real-valued function defined on the interval [a,b] and let a = x0 < x1 < x2 < … < xn = b be any
partition of [a,b].
By the variation of f over the partition P = {x 0, x1,…, xn} of [a,b], we mean the real number V(f, P)
= −
and then
Vab(f) = sup {V(f,P) for all possible partitions P of [a,b] }
= −
is called the total variation of f over the interval [a,b]. If Vab(f) < then we say that f is a function of
bounded variation and we write f ∈ BV.
4.4. Lemma. Every non-decreasing function f defined on the interval [a,b] is of bounded variation with
total variation
Vab(f) = f(b) − f(a).
Prof. For every partition P = [x0, x1, …, xn} of [a,b], we have
V(f, P) = − = −
= f(b) - f(a)
This implies the lemma.
4.5.Theorem. (Jordan Decomposition Theorem). A function f: [a,b] → R is of bounded variation if and
only if it is the difference of two non-decreasing functions.
Proof. Let f = g-h on [a,b] with g and h increasing. Then for any, subdivision we have
− ≤ − + −
= g(b) − g(a) + h(b) – h(a)
Hence,
Vab(f) ≤ g(b) − g(a) + h(b) – h(a),
which proves that f is of bounded variations.
On the other hand, let f be of bounded variation. Define two functions g, h : [a, b] →R by taking
g(x) = Vax(f), h(x) = Vax(f) − f(x) for every x ∈ [a, b]. Then f(x) = g(x) − h(x).
The function g is clearly non-decreasing. On the other hand, for any two real numbers x and y in [a, b]
with x ≤ y, we have
h(y)−h(x) = [Vax(f) − f(y)] − [Vax(f) − f(x)]
80 Measure and Integration Theory
= Vx y (f) – [f(y) − f(x)]
≥ Vx y (f) −Vx y (f) = 0
Hence h is also non-decreasing. This completes the proof of the theorem.
4.6. Examples. (1) If f is monotonic on [a,b], then f is of bounded variation on [a, b] and
V(f) = |f(b)-f(a)| , where V(f) is the total variation.
(2) If ́ (x) exists and is bounded on [a, b], then f is of bounded variation. For if |́ (x)| ≤ M we have
− ≤ − =M(b-a)
no matter which partition we choose.
(3) f may be continuous without being of bounded variation. Consider
< ≤
f(x) =
=
Let us choose the partition which consists of the points
0, , , , …… , ,
1 3 5
Then the sum in the total variation is
( 2 + ) + ( + ) + …..+ ( + )+ > + +…..+
3 1 1
and this can be made arbitrarily large by taking n large enough, since diverges.
(4). Since |f(x)− f(a)| ≤ V(f) for every x on [a,b] it is clear that every function of bounded variation is
bounded.
The Differentiation of an Integral
Let f be integrable over [a,b] and let
x
F(x) = a tdt
If f is positive, h > 0 , then we see that
xh
F(x+h) − F(x) = x tdt
Hence, integral of a positive function is non-decreasing.
We shall show first that F is a function of bounded variation. Then, being function of bounded variation,
it will have a finite differential coefficient F’ almost everywhere. Our object is to prove that ́ (x) = f(x)
almost everywhere in [a,b]. We prove the following lemma :
4.7. Lemma. If f is integrable on [a,b], then the function F defined by
x
F(x) = a tdt
is a continuous function of bounded variation on [a,b].
Proof. We first prove continuity of F. Let x0 be an arbitrary point of [a,b]. Then
Differentiation and Integration 81
x
|F(x) − F(x0)| = | tdt |
0
x
≤ ∫ tdt
0
Now the integrability of f implies integrability of |f| over [a,b]. Therefore, given > 0 there is a
> such that for every measurable set A ⊂[a,b] with measure less than , we have A < .Hence
|F(x)-F(x0)| < whenever |x−x0| < 1
and so f is continuous.
To show that F is of bounded variation, let a = x0 < x1 < … < xn = b be any partition of [a,b]. Then
− = a tdt − a 1 tdt
= tdt
1
≤ tdt
1
b
=a tdt
b
Thus Vab(f) ≤ a tdt <
Hence F is of bounded variation.
4.8. Lemma. If f is integrable on [a, b] and
x
a tdt = 0
for all x ∈ [a,b], then f = 0 almost everywhere in [a,b].
Proof. Suppose f > 0 on a set E of positive measure. Then there is a closed set F ⊂ E with m F > 0. Let
O be the open set such that
O = (a, b) – F
b
Then either a ≠ 0 or else
b
0 = a = A + O
=F + ∞
tdt (1)
because O is the union of a countable collection {(an, bn)} of open intervals.
But, for each n,
tdt = a tdt − a tdt
= F( ) –F( ) = 0 (by hypothesis) Therfore, from (1), we have
82 Measure and Integration Theory
∫ =
F
But since f > 0 on F and mF > 0, we have F > .
We thus arrive at a contradiction. Hence f = 0 almost everywhere.
4.9. Lemma. If f is bounded and measurable on [a, b] and
x
F(x) = F tdt + F(a),
xh x
then F’(x) = f(x) for almost all x in [a,b]. = a tdt − a tdt
ℎ
Proof. We know that an integral is of bounded variation over [a,b] and so F’(x) exists for almost all x in
[a,b]. Let |f| ≤ K. We set
FxhFx
=
h
with h = . Then we have
xh x
= a tdt − a tdt
ℎ
xh
= x tdt
ℎ
xh xh
implies | = x tdt ≤ x tdt
ℎ ℎ
xh
≤ x dt
ℎ
K
= .h=K
h
Moreover,
→
Hence by the theorem of bounded convergence, we have
a d = lim a d = lm ℎ a + − d
ℎ→
h
= lm ah d − a d]
ℎ→ ℎ ℎ
h ah
= lm d − a d]
ℎ→ ℎ ℎ
= F(c) –F(a)
=a d
Hence,
a − d = 0
Differentiation and Integration 83
For all c∈[a,b], and so
F’(x)= f(x) a.e.
By using pervious lemma.
Now we extend the above lemma to unbounded functions.
4.10. Theorem. Let f be an integrable function on [a,b] and suppose that
x
F(x) = F(a) +a dt
Then F’(x) = f(x) for almost all in x in [a,b].
Proof. Without loss of generality we may assume that f ≥ (or we may write “From the definition of
integral it is sufficient to prove the theorem when f ≥ 0).
Let fn be defined by fn(x) = f(x) if f(x) ≤ n and fn(x) = n if f(x) > n. Then f-fn ≥ and so
x
Gn(x) = a −
is an increasing function of x, which must have a derivative almost everywhere and this derivative will
be non-negative. Also by the above lemma, since fn is bounded (by n), we have
x
a = a.e.
Therefore,
x x
F’(x) = a = G + a
x
= G + a ≥ a.e. (using (i))
Since n is arbitrary, making n→ we see that
F’(x) ≥ f(x) a.e.
Consequently,
b b
a d ≥ a d = F(b) − F(a) (using the hypothesis of the theorem)
Also since F(x) is an increasing real valued function on the interval [a,b], we have
b b
a d ≤ F(b) —F(A) = a d
Hence
b b
a d = F(b) —F(A) = a d
b
implies a − d =
Since F’(x) − f(x) ≥0 , this implies that F’(x)− f(x) = 0 a.e. and so F’(x) = f(x) a.e.
84 Measure and Integration Theory
Absolute Continuity
4.11. Definition. A real-valued function f defined on [a,b] is said to be absolutely continuous on [a,b]
if, given > 0 there is a > 0 such that
∑ − <
for every finite collection {(xi, xi’)} of non-overlapping intervals with
∑ − <
An absolutely continuous function is continuous, since we can take the above sum to consist of one term
only. Moreover, if
x
F(x) =a tdt
Then
x x
− = a i tdt − a i tdt
x
= x i tdt
i
x
≤ x i tdt = tdt ,
i
where E is the set of intervals (x,xi’) ≤ → as − →
The last step being the consequence of the result.
“Let > 0 . Then there is a > 0 such that for every measurable set E ⊂ [a, b] with
m E < , we have A < ”.
Hence every indefinite integral is absolutely continuous.
4.12. Lemma. If f is absolutely continuous on [a,b], then it is of bounded variation on [a,b].
Proof. Let be a positive real number which satisfies the condition in the definition for = 1. Select a
natural number
n>
Consider the partition = {x0, x1,…, xn} of [a,b] defined by
x i = x0 +
for every i = 0, 1,…, n. Since |xi − xi-1| < , it follows that
Vxi1 xi (f) < 1.
This implies
Differentiation and Integration 85
Vab(f) = Vxi1 xi < n
Hence f is of bounded variation.
4.13. Corollary. If f is absolutely continuous, then f has a derivative almost everywhere.
Proof: Since f is absolutely continuous, then by above theorem, f is of bounded variation and hence f
has a derivative almost everywhere (by theorem 4.3).
4.14. Lemma. If f is absolutely continuous on [a,b] and f’(x) = 0 a.e., then f is constant.
Proof. We wish to show that f(a) = f(c) for any c ∈ [a,b].
Let E ⊂ (a,c) be the set of measure c-a in which f’(x) = 0, and let and be arbitrary positive numbers.
To each x in E there is an arbitrarily small interval [x, x+h] contained in [a,c] such that
|f(x+h) - f(x) | < h
By Vitali Lemma we can find a finite collection {[xk, yk]} of non-overlapping intervals of this sort
which cover all of E except for a set of measure less than , where is the positive number
corresponding to in the definition of the absolute continuity of f . If we label the x k so that xk ≤ xk+1,
we have (or if we order these intervals so that)
a = y0 ≤ x1 < y1 ≤ x2 < …. < yn ≤ xn+1 = c
and
− <
Now, − < − < −
by the way to intervals {[xk, yk]} were constructed, and
∑ − <
by the absolute continuity of f. Thus
f(c) − f(a) = − + − ≤ + − Since
and are arbitrary positive numbers, f(c) − f(a) = 0 and so f(c) = f(a).
Hence f is constant.
4.15. Theorem. A function F is an indefinite integral if and only if it is absolutely continuous.
Proof. Let function F is an indefinite integral then
x
F(x) = a dt
where f is integrable on [a, b]
Now f is integrable on [a, b]
is integrable on [a, b]. Then for given > , there is a > such that for every measurable
86 Measure and Integration Theory
set A contained in [a, b] with m(A) < , we have <
Let {(xi, xi’)}ni=1 be any finite collection of pairwise disjoint interval in [a, b] such that − <
Let A = ,
Then m(A) = − <
Therefore we have <
i.e., ′ <
1 ,
=> < …(1)
′
| − |
′
= | |
Consider − =
′
≤
<
Conversely, Suppose F is absolutely continuous on [a,b]. Then F is of bounded variation and we may
write
F(x) = F1(x) − F2(x),
where the functions Fi are monotone increasing. Hence F’(x) exists almost everywhere and |F’(x)| ≤
F1’(x) + F2’(x)
Thus ’d ≤ F1(b) + F2(b) − F1(a) − F2(a)
and F’(x) is integrable. Let
x
G(x) = a tdt
Then G is absolutely continuous and so is the function f = F−G. But by the above lemma since
f’(x) = ’– G’= 0 a.e., we have f to be a constant function. That is,
F(x) − G(x) = A (constant)
or
x
F(x) = a tdt = A
or
x
F(x) = a tdt + A
Taking x = a, we have A = F(a) and so
x
F(x) = a tdt + F(a)
Thus F(x) is indefinite integral of ’.
You might also like
- Full Download Probability and Measure 4th Edition Patrick Billingsley PDF100% (14)Full Download Probability and Measure 4th Edition Patrick Billingsley PDF70 pages
- Measures Integrals and Martingales 2nd Edition Ren L. Schilling Ebook All Chapters PDF100% (4)Measures Integrals and Martingales 2nd Edition Ren L. Schilling Ebook All Chapters PDF48 pages
- Lecture Notes in Measure Theory and IntegrationNo ratings yetLecture Notes in Measure Theory and Integration40 pages
- Generalised Linear Models and Bayesian StatisticsNo ratings yetGeneralised Linear Models and Bayesian Statistics35 pages
- Measure Theory and Lebesgue Integration: Appendix DNo ratings yetMeasure Theory and Lebesgue Integration: Appendix D14 pages
- Real Analysis and Infinity 1st Edition H Sedaghat 2024 scribd download100% (1)Real Analysis and Infinity 1st Edition H Sedaghat 2024 scribd download66 pages
- Discrete-Time Stochastic Process Group Work Project100% (1)Discrete-Time Stochastic Process Group Work Project11 pages
- Chapter B Probability Via Measure Theory: 1 Measurable Spaces100% (1)Chapter B Probability Via Measure Theory: 1 Measurable Spaces54 pages
- Topic #10 Markowitz Portfolio Theory: Reading: Luenberger Chapter 6, Sections 6 - 10100% (1)Topic #10 Markowitz Portfolio Theory: Reading: Luenberger Chapter 6, Sections 6 - 1093 pages
- A Brief Introduction To Measure Theory and Integration. Bass, Richard F. 1998No ratings yetA Brief Introduction To Measure Theory and Integration. Bass, Richard F. 199822 pages
- Basic Lebesgue Measure Theory: Royden 2010 Kolmogorov and Fomin 1970 Stein and Shakarchi 2005 Tao 2011No ratings yetBasic Lebesgue Measure Theory: Royden 2010 Kolmogorov and Fomin 1970 Stein and Shakarchi 2005 Tao 201128 pages
- N and F, Then (F: T.K.Subrahmonian MoothathuNo ratings yetN and F, Then (F: T.K.Subrahmonian Moothathu18 pages
- Foundations of Mathematical Analysis 1st Edition Richard Johnsonbaugh download100% (2)Foundations of Mathematical Analysis 1st Edition Richard Johnsonbaugh download51 pages
- Markov Processes - Characterization and ConvergenceNo ratings yetMarkov Processes - Characterization and Convergence273 pages
- EC744 Lecture Note 6 Stochastic Models: Mathematical PreliminariesNo ratings yetEC744 Lecture Note 6 Stochastic Models: Mathematical Preliminaries18 pages
- Election Predictions As Martingales - An Arbitrage Approach - Nassim Nicholas TalebNo ratings yetElection Predictions As Martingales - An Arbitrage Approach - Nassim Nicholas Taleb5 pages
- Further Topics On Discrete-Time Markov Control ProcessesNo ratings yetFurther Topics On Discrete-Time Markov Control Processes285 pages
- Kit-Wing Yu - A Complete Solution Guide To Real and Complex Analysis I-978-988-78797-9-4 (2019)No ratings yetKit-Wing Yu - A Complete Solution Guide To Real and Complex Analysis I-978-988-78797-9-4 (2019)330 pages
- Instant Access To Ergodic Theory Geometry and Dynamics 1st Edition C Mcmullen Ebook Full Chapters100% (7)Instant Access To Ergodic Theory Geometry and Dynamics 1st Edition C Mcmullen Ebook Full Chapters69 pages
- (H.S. Carslaw) Introduction To The Theory of Fourier SeriesNo ratings yet(H.S. Carslaw) Introduction To The Theory of Fourier Series332 pages
- Exercises in Functional Analysis: Peter SJ Ogren November 2, 2010No ratings yetExercises in Functional Analysis: Peter SJ Ogren November 2, 20106 pages
- Lecture Notes On Stochastic Calculus (NYU)100% (3)Lecture Notes On Stochastic Calculus (NYU)138 pages
- Foundations of Modern Analysis - Avner FriedmanNo ratings yetFoundations of Modern Analysis - Avner Friedman129 pages
- Abstract Algebra An Introduction 3rd Edition Thomas W. Hungerford - Read the ebook online or download it for the best experience100% (1)Abstract Algebra An Introduction 3rd Edition Thomas W. Hungerford - Read the ebook online or download it for the best experience51 pages
- Unitary and Orthogonal Operators and Their Matrices-Pt2100% (1)Unitary and Orthogonal Operators and Their Matrices-Pt222 pages
- (Ian W. Knowles, Roger T. Lewis, International Con (BookFi)No ratings yet(Ian W. Knowles, Roger T. Lewis, International Con (BookFi)401 pages
- Mathematical Conversations: Multicolor Problems, Problems in the Theory of Numbers, and Random WalksFrom EverandMathematical Conversations: Multicolor Problems, Problems in the Theory of Numbers, and Random Walks5/5 (1)
- Syllabus EEE4018 Advanced Control TheoryNo ratings yetSyllabus EEE4018 Advanced Control Theory2 pages
- 2010 Preliminary Examination II: Pre-University 3No ratings yet2010 Preliminary Examination II: Pre-University 37 pages
- Tugas Biostatistik Dosen Pengampu: Ns Ririn Afrian Sulistyawati M.KepNo ratings yetTugas Biostatistik Dosen Pengampu: Ns Ririn Afrian Sulistyawati M.Kep10 pages
- Mathematics for Joint Entrance Examination JEE (Advanced): Calculus G. Tewani - Explore the complete ebook content with the fastest download100% (1)Mathematics for Joint Entrance Examination JEE (Advanced): Calculus G. Tewani - Explore the complete ebook content with the fastest download56 pages
- Test Bank for Nursing Research Methods and Critical Appraisal for Evidence Based Practice 8th Edition by LoBiondo Wood - Available With All Chapters For Instant Download100% (13)Test Bank for Nursing Research Methods and Critical Appraisal for Evidence Based Practice 8th Edition by LoBiondo Wood - Available With All Chapters For Instant Download41 pages
- Instructional Manual For Arch 541 Architectural Design 9 1No ratings yetInstructional Manual For Arch 541 Architectural Design 9 17 pages
- Definitions: Figure 1. The Normal Distribution GraphNo ratings yetDefinitions: Figure 1. The Normal Distribution Graph8 pages
- Quantitative Techniques For Managerial DecisionsNo ratings yetQuantitative Techniques For Managerial Decisions3 pages
- Properties of Graphs of Functions: Investigate The MathNo ratings yetProperties of Graphs of Functions: Investigate The Math1 page
- One-Dimensional Assembly Tolerance Stack-Up100% (2)One-Dimensional Assembly Tolerance Stack-Up26 pages
