Scribd
Upload
192 views3 pages

HW4 Sol

The document contains solutions to problems from a real analysis homework. Problem 9 asks about the measurability of the set of points where a sequence of measurable functions converges. The solution shows this set is measurable by writing it as a countable intersection of measurable sets. Problem 11 asks about the measurability of the composition of a measurable function with a Lipschitz function, and the solution shows this composition is also measurable. Problem 22 proves Dini's theorem about uniform convergence of an increasing sequence of continuous functions.

Uploaded by

VALENTINA BALLADARES SILVA
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
192 views3 pages

HW4 Sol

The document contains solutions to problems from a real analysis homework. Problem 9 asks about the measurability of the set of points where a sequence of measurable functions converges. The solution shows this set is measurable by writing it as a countable intersection of measurable sets. Problem 11 asks about the measurability of the composition of a measurable function with a Lipschitz function, and the solution shows this composition is also measurable. Problem 22 proves Dini's theorem about uniform convergence of an increasing sequence of continuous functions.

Uploaded by

VALENTINA BALLADARES SILVA
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/ 3

Real Analysis HW 4 Solutions

Problem 9: Let {fn } be a sequence of measurable functions defined on a measurable set


E. Define E0 to be the set of points x in E at which {fn (x)} converges. Is the set E0
measurable.
Solution: Note that we may write E0 as
[
\ \
E0 = {x : {fn (x)} is Cauchy} = {x : |fn (x) fm (x)| < 1/k}.
k=1 N =1 n,mN

Therefore E0 is measurable since {x : |fn (x) fm (x)| < 1/k} is. 


Problem 11: Let f be a measurable function and g be a one-to-one function from R onto
R which has a Lipschitz inverse. Show that the composition f g is measurable.
Proof: We know from problem 38 in the previous homework that a Lipschitz function maps
measurable sets to measurable sets. Therefore to show that f g is invertible, we simply
need to show that
(f g)1 (O) = g 1 (f 1 (O))
is measurable for any open set O R. However by the measurability of f , we know that
f 1 (O) is measurable, and since g 1 is Lipschitz, then g 1 (f 1 (O)) will be measurable. 
Problem 22: (Dinis Theorem) Let {fn } be an increasing sequence of continuous functions
on [a, b] which converge pointwise on [a, b] to the continuous function f on [a, b]. Show that
the convergence is uniform on [a, b].
Solution: Let  > 0, and define En = {x [a, b] : f (x) fn (x) < }. Note that since
fSn and f are continuous and converge pointwise, that En are relatively open in [a, b] and

n=1 En = [a, b]. Furthermore, since fn is an increasing sequence then En En+1 . It follows
by Heine Borel that [a, b] is compact and therefore there exists a sub-cover {Enj }N
j=1 of [a, b].
However since En are ascending,
N
[
[a, b] = Enj = En ,
j=1

where n = maxj {nj }. Therefore the convergence is uniform. 

1
Problem 25: Suppose f is a function that is continuous on a closed set F of real numbers.
Show that f has a continuous extension to all of R. This is a special case of the forthcoming
Tietze Extension Theorem.
Solution: We may assume F is non-empty. Since F is closed we see that R F is open,
S therefore may be written as a countable disjoint union of open intervals R F =
and
k=1 (ak , bk ). On each of these intervals, we extend f continuously to f as a linear function
in the following way: suppose x (ak , bk ), then

if ak , bk are finite we define

f (bk ) f (ak )
f(x) = (x ak ) + f (ak ),
b k ak

if bk = define
f(x) = f (ak ),

if ak = define
f(x) = f (bk ).

Clearly the extended f is now continuous on R since it matches f on the endpoints of the
intervals and is therefore continuous at every point. 
Problem 27: Show that the conclusion of Egoroffs Theorem can fail if we drop the as-
sumption that the domain has finite measure.
Solution: Consider sequence fn (x) = [n,) (x). Clearly fn 0 pointwise on R. Suppose
that there existed a set F such that m(R F ) <  and fn 0 uniformly on F . Since fn is
an indicator function, uniform convergence means we can find an N such that fN = 0 on F .
This implies F (, N ) and so [N, ) R F , giving

m([N, )) m(R F ) < ,

which is a contradiction. 
Problem 29: Prove the extension of Lusins Theorem to the case that E has infinite
measure.
Solution: Suppose f : E R is measurable. We split up E into disjoint finite measure
sets {En }nZ given by En = E [n, n + 1). By Lusins theorem for finite measure sets,
S functions gn : Fn R such that
we know that there exist closed sets Fn and continuous
m(En Fn ) < /2|n|+1 and f = gn on Fn . Define F = kZ Fn and
X
g(x) = gn (x)Fn (x).
kZ

Note that g is continuous on F .

2
We now claim that F is closed. To see this consider {xn } F , converging to an x E.
Since x Ek for some k, we see that for large enough N , {xn }nN belongs to Fk1 Fk
which is a closed set and therefore x Fk1 Fk F .
Since F is closed, by problem (25) we may extend g continuously to R. It now follows that
f = g on F and
!
[ X
m(E F ) = m (En Fn ) = m(En Fn ) < .
nZ kZ

You might also like