MEASURE AND INTEGRATION: LECTURE 5
Denition of L1 . Let f : X [, ] be measurable.
We say that
f is in L1 (written f L1 () or simply f L1 ) X f + d <
and X f d X |f | d < . Dene
+
f d =
f d
f d
X
when at least one of the terms on the righthand side is nite.
Integral of complex functions. Let f : X C be measurable. That
is, f = u + iv where u, v : X
R are measurable. Then
1
f L ()
|F | d <
|u| d < and
|v | d < .
X
Dene
f d =
u d + i
v d
X
+
u d
v d i
u d + i
v d.
Theorem 0.1. Let f, g L (). If , C, then f + g L1 ()
and
(f + g)d =
f d +
g d.
X
Proof. First, f + g is measurable, and by the triangle inequality,
|f + g| || |f | + || |g |, so
|f + g| ||
|f | + ||
|g | < .
X
Just need to show that
(1) X (f + g)d = X f d + X g d, and
(2) X (f )d = X f d.
Date: September 18, 2003.
1
MEASURE AND INTEGRATION: LECTURE 5
For (1), assume f, g real; the complex case follows from the real case.
Let h = f + g. Then h+ h = f + f + g + g , so h+ + f + g =
f + + g + + h . Since the integral is linear for nonnegative functions,
+
+
h + f + g = f + g + h
+
h h = f f + g g.
Thus, f + g = f + g.
For (2), let = a + bi for a, b R. Then
f = (a + bi)(u + iv) = au + aiv + biu bv
= (au bv + i(av + bu)
= (au bv) + i (av + bu).
Also,
(a + bi)
(u + iv) = (a + bi)
u+i v
= a u + bi u + ai v b v.
So, just need to show that au = a u. If a = 0, then both sides
vanish. If a > 0, then
+
(au) = (au) (au)
+
= a u a u
= a u a u = a u.
If a < 0, then
(au)
= (a) u (a) u+
= a u (a) u+
=a
u u
= a u.
au =
(au)
MEASURE AND INTEGRATION: LECTURE 5
Theorem 0.2. If f L1 (), then
f d
|f | d.
Proof. For some [0, 2),
f d =
f d
ei .
Hence,
f d = ei
f d =
(ei f )d
X
X
X
i
= Re
e f d
X
=
Re(ei f )d
X
e f d =
|f | d.
X
Dominated convergence.
Theorem 0.3. Let fn : X C be a sequence of measurable functions,
and assume that f (x) = limn fn (x) (that is, the sequence fn con
verges pointwise). If there exists g L1 () such that |fn (x)| g(x)
for all n and for all x X, then f L1 () and
lim
|fn f | d = 0, so lim
fn d =
f d.
n
Proof. Since |fn (x)| g(x) for alln, the limit
f has the property 1that
|f | |g(x)|. This means that |f | |g | < ,
so f L ().
Next, |fn f | |fn | + |f | 2g, which means that 2g |fn f | 0.
Applying Fatous lemma,
2g d lim inf (2g
|fn f |)d
X
X
=
2g d + lim inf
|fn f | d
X
X
=
2g d lim sup
|fn f | d.
X
MEASURE AND INTEGRATION: LECTURE 5
Since
2g < , it can be cancelled from both sides. Thus,
lim sup
|fn f | d 0,
X
and so
|fn f | d = 0.
lim
X
From the previous theorem,
(fn f )d
|fn f | d
X
X
fn d
f d
|fn f | d 0
X
X
X
fn d
f d.
Sets of measure zero. Let (X, M, ) be a measure space and E
M. A set E has measure zero if and only if (E) = 0. If f, g : X C,
then f = g almost everywhere (a.e.) if N = {x | f (x) = g(x)} has
measure zero. Dene an equivalence relation f g if f = g a.e.
Proposition 0.4. If f g, then, for all E M, E f d = E g d.
Proof. Write E as disjoint union E = (E \ N ) (E N ). Then, since
f = g away from N , and since N has measure zero,
f d =
f d +
f d
E
E\N
EN
=
g d + 0 =
g d.
E\N
Completion of a algebra.
Theorem 0.5. Let (X, M, ) be a measure space. Let
M = {E X | A, B M : A E B & (B \ A) = 0}.
Now dene (E) = (A) for all E M . Then M is a algebra and
this denition of is a measure.
The measure space (X, M , ) is a called the completion of the mea
sure space (X, M, ). A measure space is complete if it is equal to its
completion.
Note.
If f is only dened a.e. (say, except for aset N of measure
zero), then we can dene f (x) = 0 for all x N . f is well dened.
MEASURE AND INTEGRATION: LECTURE 5
Theorem
(a) Let f : X [0, ] be measurable, E M, and
0.6.
f
d
=
0.
Then f = 0 a.e. on E.
E
1
(b) Let f L () and E f d = 0 for every E M. Then f = 0
a.e. on X.
Proof.
(a) Let An = {x E | f (x) > 1/n}. Then
1
f d
f d
1/n d = (An ),
n
E
An
An
which implies that (An )
= 0. But {x | f (x) > 0} =
i=1 An
and ({x | f (x) > 0})
(A
)
=
0.
n
i=1
(b) Let
f
=
u
+
iv.
Choose
E
=
{
x
|
u
(
x
)
0
}
.
Then
f =
E
+
+
+
u + i E v E u = 0 and by (a), u = 0 a.e.
E
Theorem 0.7. Let Ek M such that k=1 (Ek ) < . Then almost
every x X lie in at most nitely many Ek .
Proof. Let
A = {x X | x Ek for innitely many k}. NTS (A) = 0.
Let g =
i=1 Ek .
Then x A g(x) = . We have
g d =
Ek d =
(Ek ) < .
X
i=1
1
i=1
In other words, g L () and thus g(x) < a.e.