Real Analysis HW 1 Solutions
Problem 1: Prove that if A and B are two sets in A with A B, then m(A) m(B).
Proof: Since A B we can split up B into a union of two disjoint sets B = A (B A).
Using the countable additivity of m we find
m(B) = m (A (B A)) = m(A) + m(B A) m(A),
since m(B A) 0.
Problem
3: Let {Ek }
k=1
P
m(E
).
k
k=1
S
be a countable collection of sets in A. Prove that m( k=1 Ek )
Proof: Using the collection
{Ek }
k=1Swe construct a countable disjoint collection of sets
S
{Ak }k=1 A such that k=1 Ak =
k=1 Ek in the following way. Choose A1 = E1 and
define for n > 1,
n1
[
Ek .
An = En
k=1
S
S
It is easy to see that {An }
n=1 A are pairwise disjoint,
k=1 Ak , and Ak Ek .
k=1 Ek =
Therefore using countable additivity and monotonicity,
!
!
X
X
[
[
m(Ek ).
m(Ak )
Ak =
m
Ek = m
k=1
k=1
k=1
k=1
Problem 5: By using the properties of an outer measure, prove that the interval [0, 1] is
not countable.
Solution: We know the outer measure of an interval is its length, so m ([0, 1]) = 1,
howevever we also know that the outer measure of a countable set if 0. Therefore [0, 1]
cannot be countable.
Problem 8: Let B be the set of rational numbers in the P
interval [0, 1],and let {Ik }nk=1 a
finite collection of open intervals that covers B. Prove that nk=1 m (Ik ) 1.
Solution: Since the rationals are dense in R we see that the closure B = [0, 1]. Therefore
S
[0, 1] nk=1 Ik . However, since the closure of the union is just the union of the closure
1
S
S
for finite
unions
(note
that
in
general
for
countable
unions
I
k
k=1
k=1 Ik ), we have
Sn
B k=1 Ik . Therefore by the finite subadditivity of m ,
!
n
n
n
[
X
X
1 = m ([0, 1]) = m (B) m
Ik
m (Ik ) =
m (Ik ),
k=1
k=1
k=1
where we have used the fact that Ik are intervals, so m (Ik ) = l(Ik ) = m (Ik ).
Problem 11: Prove that if a -algebra of subsets of R contains intervals of the form (a, ),
then it contains all intervals.
Solution: Let be the -algebra. By complement must contain intervals of the form
(, a]. Also intervals
of the form (, a) are in since they can be given by a coutable
S
union (, a) =
(,
a 1/n] and so by complement [a, ) is also in . We conclude
k=1
that for any a, b finite,
(a, b) = (, b) (a, )
(a, b] = (, b] (a, )
[a, b) = (, b) [a, )
[a, b] = (, b] [a, )
Therefore contains all intervals.
Problem 14: Show that if a set E has positive outer measure, then there is a bounded
subset of E that also has positive outer measure.
Proof: We prove by contradiction. Suppose that every bounded subset of E has outer
measure zero. Define Ik = [k, k +1], to be a countable collection of disjoint bounded intervals
that decompose R. We can then decompose E as a countable union of bounded subsets of
E
[
E Ik ,
E=
kZ
and so by the hypothesis m (E Ik ) = 0. Therefore by finite sub-additivity of m ,
!
[
X
0 < m (E) = m
E Ik
m (E Ik ) = 0
kZ
kZ
which is a contradiction.
Problem 15: Show that if E has finite measure and > 0, then E is the disjoint union of
a finite number of measurable sets, each of which has measure at most .
Solution: Let Bk = (k/2, k/2) and define L1 = B1 , Lk = Bk Bk1 for k 2. Note
that {Lk }
k=1 are disjoint and m (Lk ) = . Now, define the countable disjoint collection of
measurable sets {Ek }
k=1 , where Ek = E Lk . Since E has finite measure, by countable
additivity
X
m (Ek ) = m (E) < .
k=1
Since this series is convergent, there exists an N such that
E0 =
k=N +1
m (Ek ) < . Define
Ek ,
k=N +1
then m(E0S
) < . It follows that {Ek }N
k=0 is a finite disjoint collection of measurable sets such
that E = N
E
and
m
(E
)
<
.
k
k=0 k
