MEASURE AND INTEGRATION: LECTURE 17

Inclusions between Lp spaces. Consider Lebesgue measure on the

space (0, ) R. Recall that xa is integrable on (0, 1) a > 1,
and it is integrable on (1, ) a < 1. Now let 1 p < q .
Choose b such that 1/q < b < 1/p. Then xb (0,1) is in Lp but not in
Lq , which shows that Lp Lq . On the other hand, xb (1,) is in Lq
but not in Lp , so that Lq Lp . Thus, in general there is no inclusion
relation between two Lp spaces.
The limit of f p as p . For convenience, dene f p to be
if f is Mmeasurable but f Lp .
Theorem 0.1. Let f Lr for some r < . Then
lim f p
= f .

This justies the notation for the L norm.

Proof. Let t [0, f ). By denition, the set

A = {x X | |f (x)| t}
has positive measure. Observe the trivial inequality

1/p
p
f p
|f | d
A

(t (A))1/p
p

= t(A)1/p .
If (A) is nite, then (A)1/p 1 as p . If (A) = , then
(A)1/p = . In both cases, we have
lim inf f p t.
p

Since t is arbitrary,
lim inf f p f .
p

Date: October 30, 2003.

1

MEASURE AND INTEGRATION: LECTURE 17

For the reverse inequality, we need the assumption that f Lr for

some (nite) r. For r < p < , we have
1r/p
f p f r/p
.
r f

Since f r <
,
lim sup f p f .

The inequality used in the proof can be written as

f p p
({x X | |f (x)| t})
,
t
and is known as Chebyshevs inequality.
Finite measure spaces. If the measure of the space X is nite, then
there are inclusion relations between Lp spaces. To exclude trivialities,
we will assume throughout that 0 < (X) < .
Theorem 0.2. If q p < q < , then Lq Lp .
Proof. Applying Holders inequality to |f |p and 1,

p
|f | d = |f |p 1 d

pq/p

|f |

p/q
1p/q
d
d

p/q
|f | d
(X)1p/q .
q

In particular, if (X) = 1, then

f 1 f p f q f .
Counting measure and lp spaces. Let X be any set, M = P(X),
and be the counting measure. Recall that (A) is the number of
points in A if A is nite and equals otherwise. Integration is simply

f d =
f (x)
X

xX
p

for any nonnegative function f , and L is denoted by lp .

Theorem 0.3. If 1 p < q , then lp lq , and
f f q f p f 1 .

MEASURE AND INTEGRATION: LECTURE 17

Proof. If q = , then observe that for any x0 X,

1/p

p
| f (x0 )|
|f (x)|
.
xX

Now let q < . Then we NTS

1/q
1/p

q
p
|f (x)|

|f (x)|
.
xX

xX

Now multiply both

is equal to 1.
sides pby a constant so thatthe RHS
q
Thus, assuming
|f (x)| = 1, we NTS that
|f (x)| 1. But this
is immediate, since |f (x)| 1 for all x implies that |f (x)|q |f (x)|p
because q > p.

Thus, in a certain sense, the counting measure and a nite measure

act in reverse ways for Lp spaces.
Local LP spaces. Let G be an open set in Rn . The local Lp space on
G consists of all Lmeasurable functions f dened a.e. on G such that
for every compact set K G, the characteristic function f K has a
nite Lp norm; that is,

|f (x)|p dx < if 1 p < ;

K

f is essentially bounded on K

if p = .

This set is denoted Lploc (G). From our result on nite measure spaces,
we have at once for 1 p < q ,
p
q
1
L
loc (G) Lloc (G) Lloc (G) Lloc (G).

Convexity properties of Lp norm. Let (X, M, ) be a measure

space.
Theorem 0.4. Let 1 p < r < q < and suppose f Lp Lq .
Then f Lr and
log f r

1
r
1
p

1
q
1
q

log f p +

1
p
1
p

1
r
1
q

log f q .

Proof. Since 1/q < 1/r < 1/p, there exists a unique such that
1
1
= +
.
r
p
q

MEASURE AND INTEGRATION: LECTURE 17

The number satises 0 < < 1 and equals

=

1
r
1
p

1
q
1
q

1 =

1
p
1
p

1
r
1
q

We NTS that log f r log f p + (1 ) log f q . Note that

r r(1 )
+
,
p
q
and so p/r and q/r(1 ) are conjugate exponents. Thus, by Holders
inequality,

f r =
f f 1
r

1/r
=

f r f r(1)
1
1/r

f r
p/r
f r(1)
q/r(1)
1/r

r(1)
= f r
f

p
q
1=

= f p f 1
.
q

The theorem states that if f is an Mmeasurable nonzero function

on X, then the set of indices p such that f Lp is an interval I [1, ],
and log f p is a convex function of 1/p on I.