Scribd
Upload
57 views5 pages

Introduction To Lebesgue Integration: Wwlchen

This document introduces the dominated convergence theorem, which states that if a sequence of functions converges almost everywhere to a limit function and is dominated by an integrable function, then the limit function is integrable and the integrals of the sequence converge to the integral of the limit function. The proof constructs two sequences that bound the given sequence from above and below and converge to the limit function. Several consequences of the theorem are also presented, including the bounded convergence theorem.

Uploaded by

Fabian Molina
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
57 views5 pages

Introduction To Lebesgue Integration: Wwlchen

This document introduces the dominated convergence theorem, which states that if a sequence of functions converges almost everywhere to a limit function and is dominated by an integrable function, then the limit function is integrable and the integrals of the sequence converge to the integral of the limit function. The proof constructs two sequences that bound the given sequence from above and below and converge to the limit function. Several consequences of the theorem are also presented, including the bounded convergence theorem.

Uploaded by

Fabian Molina
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

INTRODUCTION TO

LEBESGUE INTEGRATION
W W L CHEN
c

W W L Chen, 1977, 2008.

This chapter was first written in 1977 while the author was an undergraduate at Imperial College, University of London.
It is available free to all individuals, on the understanding that it is not to be used for financial gain,
and may be downloaded and/or photocopied, with or without permission from the author.
However, this document may not be kept on any information storage and retrieval system without permission
from the author, unless such system is not accessible to any individuals other than its owners.

Chapter 6
DOMINATED CONVERGENCE THEOREM

6.1. Lebesgues Theorem


In this section, we shall deduce the following result from the Monotone convergence theorem studied in
the last chapter. The result below is usually considered the cornerstone of Lebesgue integration theory.
THEOREM 6A. (LEBESGUES THEOREM) Suppose that I R is an interval. Suppose further that
the sequence of functions fn L(I) satisfies the following conditions:
(a) The sequence fn : I R converges almost everywhere to a limit function f : I R.
(b) There exists a non-negative function F L(I) such that for every n N, |fn (x)| F (x) for almost
all x I.
Then the limit function f L(I), the sequence
Z
fn (x) dx
I

is convergent, and
Z

Z
f (x) dx = lim

fn (x) dx.

(1)

Remark. Note condition (b) that the sequence fn is dominated by F almost everywhere.
Proof of Theorem 6A. We shall construct two sequences gn , hn L(I) such that
gn (x) fn (x) hn (x)
Chapter 6 : Dominated Convergence Theorem

(2)
page 1 of 5

Introduction to Lebesgue Integration

W W L Chen, 1977, 2008

for every x I, and where gn is increasing and hn is decreasing on I, and both converge to the limit
function f almost everywhere on I. Clearly the sequence
Z
gn (x) dx
I

is increasing and bounded above by


Z
F (x) dx,
I

so that
Z
lim

gn (x) dx

exists. It follows from Theorem 5C that f L(I) and


Z
Z
f (x) dx = lim
gn (x) dx.
n

(3)

On the other hand, the sequence


Z
hn (x) dx
I

is decreasing and bounded below by


Z

F (x) dx,
I

so that
Z
lim

hn (x) dx
I

exists. It follows from Theorem 5C (applied to the sequence hn ) that


Z
Z
f (x) dx = lim
hn (x) dx.
n

Combining (3) and (4), we obtain


Z
Z
Z
f (x) dx = lim
gn (x) dx = lim
hn (x) dx.
n

(5)

On the other hand, it follows from (2) that for every n N,


Z
Z
Z
gn (x) dx fn (x) dx hn (x) dx.
I

(4)

(6)

The equality (1) follows on letting n in (6) and combining with (5). It remains to establish the
existence of the sequences gn and hn . For every n N, write
hn (x) = sup{fn (x), fn+1 (x), fn+2 (x), . . .}
for every x I. Clearly fn (x) hn (x) for every x I, and hn is decreasing on I. Suppose that x I
and fn (x) f (x) as n . Then given any  > 0, there exists N such that for every n > N ,
f (x)  < fn (x) < f (x) + .
Chapter 6 : Dominated Convergence Theorem

page 2 of 5

Introduction to Lebesgue Integration

W W L Chen, 1977, 2008

It follows that for all n > N ,


f (x)  < hn (x) f (x) + ,
so that hn (x) f (x) as n . Since fn f as n almost everywhere on I, it follows that
hn f as n almost everywhere on I. Unfortunately, we also need to show that hn L(I). Here,
the difficulty arises since hn (x) is defined as the supremum of a collection which may be finite or infinite.
This difficulty would not have arisen if the collection were finite, since the supremum of such a collection
would then be equal to its maximum, and we could then use Theorem 4N repeatedly. However, the finite
case suggests the following approach. For every m, n N with m > n, write
hnm (x) = max{fn (x), fn+1 (x), . . . , fm (x)}
for every x I. Then by repeated application of Theorem 4N, we have hnm L(I). For every fixed
n N, the sequence hnm (in counting variable m > n) is increasing on I. On the other hand, clearly
|hnm (x)| F (x) for almost all x I. It follows that
Z
Z
Z


hnm (x) dx |hnm (x)| dx F (x) dx.


I

Hence the sequence


Z
hnm (x) dx
I

is increasing and bounded above and so converges. It follows from Theorem 5C that hnm converges
almost everywhere as m to a limit function in L(I). Clearly hnm hn as m . This proves
that hn L(I). Similarly, write
gn (x) = inf{fn (x), fn+1 (x), fn+2 (x), . . .}
for every x I. Clearly gn (x) fn (x) for every x I, and gn is increasing on I. Suppose that x I
and fn (x) f (x) as n . Then given any  > 0, there exists N such that for every n > N ,
f (x)  < fn (x) < f (x) + .
It follows that for all n > N ,
f (x)  gn (x) < f (x) + ,
so that gn (x) f (x) as n . Since fn f as n almost everywhere on I, it follows that
gn f as n almost everywhere on I. To show that gn L(I), for every m, n N with m > n,
write
gnm (x) = min{fn (x), fn+1 (x), . . . , fm (x)}
for every x I. Then by repeated application of Theorem 4N, we have gnm L(I). For every fixed
n N, the sequence gnm (in counting variable m > n) is decreasing on I. On the other hand, clearly
|gnm (x)| F (x) for almost all x I. It follows that
Z
Z
Z


gnm (x) dx |gnm (x)| dx F (x) dx.


I

Hence the sequence


Z
gnm (x) dx
I
Chapter 6 : Dominated Convergence Theorem

page 3 of 5

Introduction to Lebesgue Integration

W W L Chen, 1977, 2008

is decreasing and bounded below and so converges. It follows from Theorem 5C (applied to the sequence
gnm ) that gnm converges almost everywhere as m to a limit function in L(I). Clearly gnm gn
as m . This proves that gn L(I). The proof of Theorem 6A is now complete.
The following version for a series can be deduced easily from Theorem 6A.
THEOREM 6B. Suppose that I R is an interval. Suppose further that the sequence of functions
gn L(I) satisfies the following conditions:

X
(a)
gn converges almost everywhere on I to a sum function g : I R.
n=1
N

X



(b) There exists a non-negative function G L(I) such that for every N N,
gn (x) G(x) for


n=1
almost all x I.
Then g L(I), the series
Z
X

gn (x) dx

n=1

converges, and
Z
g(x) dx =
I

Z X

gn (x) dx =

I n=1

Z
X
n=1

gn (x) dx.

6.2. Consequences of Lebesgues Theorem


The following result is sometimes called the Bounded convergence theorem.
THEOREM 6C. Suppose that I R is a bounded interval. Suppose further that the sequence of
functions fn L(I) satisfies the following conditions:
(a) The sequence fn : I R converges almost everywhere to a limit function f : I R.
(b) There exists M R such that for every n N, |fn (x)| M for almost all x I.
Then the limit function f L(I), and
Z

Z
f (x) dx = lim

fn (x) dx.
I

Remark. In view of conditions (a) and (b), we say that the sequence fn is boundedly convergent almost
everywhere on I.
Proof of Theorem 6C. Let F (x) = M for every x I, and note that since I is a bounded interval,
we have F L(I). The result now follows from Theorem 6A.
The last result in this section is sometimes useful in establishing Lebesgue integrability.
THEOREM 6D. Suppose that I R is an interval. Suppose further that the sequence of functions
fn L(I) satisfies the following conditions:
(a) The sequence fn : I R converges almost everywhere to a limit function f : I R.
(b) There exists a non-negative function F L(I) such that |f (x)| F (x) for almost all x I.
Then the limit function f L(I).
Chapter 6 : Dominated Convergence Theorem

page 4 of 5

Introduction to Lebesgue Integration

W W L Chen, 1977, 2008

Proof. For every n N, write


gn (x) = max{min{fn (x), F (x)}, F (x)}
for every x I (the reader is advised to draw a picture). Then gn L(I) by Theorem 4N. It is easy to
see that |gn (x)| F (x) for almost all x I, and that gn f as n almost everywhere on I. The
result follows from Theorem 6A.

Chapter 6 : Dominated Convergence Theorem

page 5 of 5

You might also like