Real Analysis HW2
Real Analysis HW2
2012311871
Jaemin Shin
Real Analysis HW # 2
Chapter 9
Exercise 2. Prove Youngs theorem (9.2).[For f, g 0 and p, q, r < , write
(f g)(x) =
f(t)
p/r
g(x t)
q/r
f(t)
p(1/p1/r)
g(x t)
q(1/q1/r)
dt,
and apply Holders inequality for three functions (Exercise 6, Chapter 8) with exponents r, p
1
, and p
2
,
where 1/p
1
= 1/p 1/r, 1/p
2
= 1/q 1/r.]
Theorem 1 (9.2). Let p and q satisfy 1 p, q and
1
p
+
1
q
1, and let r be dened by
1
p
+
1
q
= 1+
1
r
.
If f L
p
(R
n
) and g L
q
(R
n
), then f g L
r
(R
n
) and
f g
r
f
p
g
q
.
proof ) By assumption we can write as
1
r
+
_
1
p
1
r
_
. .
=1
1
q
+
_
1
q
1
r
_
. .
=1
1
p
= 1.
And apply the generalized Holder inequlity
|(f g)(x)|
|f(t)g(x t)|dt
_
|f(t)|
p
|g(x t)|
q
dt
_1
r
_
|f(t)|
p
dt
_1
p
1
r
_
|g(x t)|
q
dt
_1
q
1
p
.
Raising this inequality to the rth power we get, because of the translation invariance of
Lebesque measure,
|(f g)(x)|
r
_
|f(t)|
p
|g(x t)|
q
dt
_
f
rp
p
g
rq
q
= |f|
p
|g|
q
(x) f
rp
p
g
rq
q
.
Now we integrate this inequality over x and use the integral
|f|
p
|g|
q
(x)dx = |f|
p
|g|
q
1
f
p
p
g
q
q
.
Thus,
f g
r
r
=
|f g(x)|
r
dx
f
p
p
g
q
q
f
rp
p
g
rq
q
= f
r
p
g
r
q
.
Therefore,
f g
r
f
p
g
q
.
1
Exercise 4. (a) Show that the function h dened by h(x) = e
1/x
2
for x > 0 and h(x) = 0 for x 0 is in C
.
proof ) Since the function h(x) can be expressed as the Taylor expansion when x > 0 and h(x) = 0
for x 0, then h(x) C
(b) Show that the function g(x) = h(x a)h(b x), a < b, is C
.
And by (a), g(a) = g(b) = h(0)h(0) = 0. That is, g(x) is in C
0
(R
n
) whose support is a ball or an interval.
proof )
f(x) =
_
e
1
2|x|
, |x| 2
0, |x| > 2
where, x = (x
1
, x
2
, . . . , x
n
).
Therefore, f(x) is in C
0
(R
n
) and its support is a ball.
2
Exercise 6. Prove theorem (9.4)
Theorem 2 (9.4). If f L(R
n
) and K is bounded and uniformly continuous on R
n
, then f K is
bounded and uniformly continuous on R
n
.
proof ) (i)
|f K(x)| =
K(x t)f(t)dt
sup K
|f(t)|dt
while f L(R
n
), so
|f(t)|dt < .
And K is bounded, then
|f K(x)| C.
(ii) There exist > 0 such that for all u, w, |w u| .
|f K(x) f K(y)| =
_
K(x t) K(y t
_
f(t)dt
sup
|wu|
|K(w) K(u)|f
L
1.
Since K is bounded and uniformly continuous, f K is bounded and uniformly continuous.
3
Exercise 8. (Schurs lemma) For s, t 0, let K(s, t) satisfy K 0 and K(s, t) =
1
K(s, t) for all > 0, and
suppose that
0
t
1/p
K(1, t)dt = < + for some p, 1 p . For example, K(s, t) = 1/(s + t)
has these properties. Show that if
(Tf)(s) =
0
f(t)K(s, t)dt, (f 0),
then Tf
p
f
p
. [Note that K(s, t) = s
1
K(1, t/s), so that (Tf)(s) =
0
f(ts)K(1, t)dt. Now
apply Minkowskis integral inequality (see Exercise 8, Chapter 8).]
proof ) Since K(s, t) =
1
K(s, t), then
K(1, t/s) = sK(s, t).
Thus,
(Tf)(s) =
0
f(t)K(s, t)dt =
0
f(ts)K(1, t)dt.
=
Tf
p
=
_
0
|(Tf)|
p
ds
_1
p
=
_
0
f(ts)K(1, t)dt
p
ds
_1
p
0
_
0
|f(ts)K(1, t)|
p
ds
_1
p
dt
=
0
|K(1, t)|
_
0
|f(ts)|
p
ds
_1
p
dt
=
0
t
1
p
|K(1, t)|
_
0
t|f(ts)|
p
ds
_1
p
dt
=
0
t
1
p
|K(1, t)|
_
0
|f(ts)|
p
dts
_1
p
dt
= f
p
.
4
Exercise 10. Let T : f Tf be a function transformation which is sublinear; that is, T has the property that if
Tf
1
and Tf
2
are dened, then so is T(f
1
+ f
2
), and
|T(f
1
+ f
2
)(x)| |(Tf
1
)(x)| +|(Tf
2
)(x)|.
Suppose also that there are constants c
1
and c
2
such that T satisfy Tf
c
1
f
and |{x :
|(Tf)(x)| > }| c
2
1
f
1
, > 0. Show that for 1 < p < , there is a constant c
3
such that
Tf
p
c
3
f
p
. This is a special case of an interpolation result due to Marcinkiewicz. (An example
of such a T is the maximal function operator Tf = f
.)
proof ) Let
Tf
() = |{x : |(Tf)(x)| > }| and let f be a measurable function and > 0. Denote
by f
1
and f
2
the following functions.
f
1
(x) =
_
f(x), |f(x)| /2c
1
0, otherwise
f
2
(x) =
_
f(x), |f(x)| > /2c
1
0, otherwise
.
Since Tf
c
1
f
, f = f
1
+ f
2
, and |Tf| |Tf
1
| +|Tf
2
|,
Tf
1
c
1
f
1
2
. (1)
And then from (1)
Tf
()
{x : |(Tf
1
)(x)| >
2
}
{x : |(Tf
2
)(x)| >
2
}
{x : |(Tf
2
)(x)| >
2
}
2c
2
1
f
2
1
= 2c
2
{x:|f|>/2c
1
}
|f|dx.
Since
Tf
p
p
= p
p1
Tf
()d,
1
p
Tf
p
p
p1
_
2c
2
{x:|f|>/2c
1
}
|f|dx
_
d
= 2c
2
p2
_
{x:|f|>/2c
1
}
|f|dx
_
d
= 2c
2
R
n
|f|
2c
1
|f|
0
p2
ddx
=
2c
2
(2c
1
)
p1
p 1
R
n
|f|
p
dx
=
2
p
c
2
c
p1
1
p 1
f
p
p
.
Therefore,
Tf
p
c
3
f
p
,
where
c
p
3
=
2
p
pc
2
c
p1
1
p 1
.
5
Exercise 12. Show that the conclusions of (9.9) and (9.13) remain true if the assumption that f L
1
is replaced by
f L
p
, p > 1.
proof )
6