This action might not be possible to undo. Are you sure you want to continue?
Homework Set #1
by Bobby Rohde
9700
Problem 1
If A and B are two sets in with A B, then mA mB.
Proof
Since is a oalgebra, we know that C = B A
e , with C A = O and C A = B. Hence mB
= m(C A) = mC + mA ± mA. . mB ± mA, QED.
Problem 2
Let <E
n
> be any sequence of sets in . Then m( E
n
) mE
n
.
Proof
By Proposition 1.2, ¬ a sequence <A
n
> of sets in with E
n
= A
n
and A
n
A
m
= O, ! n = m.
Thus m( E
n
) = m( A
n
) = mA
n
. If we construct A
n
from E
n
according to the algorithim used
in the proof of Proposition 1.2 then we have A
n
· E
n
! n, and from Problem 1 therefore mA
n
s
mE
n
. Hence we have mA
n
s mE
n
= m( E
n
) s mE
n
. QED.
Problem 3
If there is a set A in such that mA < , then m = 0.
Proof
By Way of Contradiction (BWOC). Suppose m = 0, and let mA = < . A = , hence
A and are disjoint. Thus by Property 3, m(A ) = mA + m = + . But A = A, so m(A
) = mA = . Thus = + > . Which is a contradiction, hence m = 0. QED.
Problem 4
Let nE be for an infinite set E and equal to the number of elements in E
for a finite set. Show that n is a countably additive set function that is
translation invariant and defined for all sets of real numbers.
Proof
Countably Additive
We need to show that disjoint sequences <A
i
> of sets in , n( A
i
) = nA
i
.
If A
i
is infinite for any i then n( A
i
) = nA
i
= , so we may as well assume that A
i
is finite i.
However since A
i
is disjoint that implies that the number of elements in A
i
A
j
= nA
i
+ nA
j
, i
j. Thus by repeated applications it is clear that n( A
i
) must equal nA
i
. QED.
Translation Invariant
We need to show that n(A + y) = nA y.
Clearly the definition of A + y = {x + y : x A}, establishes a 11 correspondance between
elements of A and elements of A + y, thus A and A + y must have the same number of elements and
hence n(A + y) = nA y. QED
Defined on All Sets of
Clearly it is in the nature of sets that they must have either an infinite number of elements or
exactly one welldefined finite number of elements so it is impossible to construct a set on which
does not have a unique value under operation by n. Thus all sets on are measurable by n. QED.
MATH6301.nb 2
Problem 5
Let A be the set of rational numbers between 0 and 1, and let {I
n
} be a
finite collection of open intervals covering A. Show that l(I
n
) 1.
Proof
Let (a
n
, b
n
) denote the interval I
n
with end points a
n
< b
n
.
BWOC Suppose 0 < a
n
, ! n, then since rationals are dense, ¬ a rational number , between 0 and
min{a
n
}, not covered by I
n
, which is a contradiction. So we know ¬ n such that a
n
s 0. Similarly
we must have an m such that b
m
± 1.
BWOC Suppose ¬ I
n
with b
n
< 1 such that ! m = n, a
m
s b
n
< b
m
is false = ¬ an open interval C =
(b
n
, min({a
n
: a
n
> b
n
} {1})), which is not in I
n
. But since the rational numbers are dense, ¬ a
rational number in C, which contradicts the fact that {I
n
} covers A. Thus ! I
n
with b
n
< 1, ¬ m
such that a
m
s b
n
< b
m
.
Consider T = ( I
n
) {b
n
} {a
n
}. Clearly T covers A since I
n
covers A. Furthermore since
a
m
s b
n
< b
m
is true under the conditions just stated, this implies that including {a
n
} and {b
n
} will
close any gaps between intervals. Since there are a finite number of open intervals, we are adding
only a finite number of points to T, and hence m*T = l(I
n
). But T is now a continuous interval
with (0, 1) c T, using the fact that ¬ n, m such that a
n
s 0 and b
m
± 1. Thus m*T ± 1  0 = 1. . l(
I
n
) ± 1. QED.
Problem 6
Prove that: Given any set A and any > 0, an open set O such that A O
and m*O m*A + . There is a G G
such that A G and m*A = m*G.
Proof
Since inf
A·I
n
lI
n
= m*A we know by definition of infimum that ¬ a collection of open intervals
{I
n
} where lI
n
s m*A + e ! e > 0. Thus we have that O = I
n
is an open set with m*O =
lI
n
s m*A + e. Which gives us the first part of the Proposition
For the second part we merely have to consider G =
n=1
o
O1
n
, where O
e
is the O associated with a
specified e in the first part. Since A · O
e
! e > 0, we know that A c G. Also since G is an
MATH6301.nb 3
intersection of a countable collection of open intervals, G e G
5
. Finally since m*A s m*O s m*A +
e, with e ÷ 0 in our construction of G, we can therefore conclude that m*A = m*G. QED.
Problem 7
Prove that m* is translation invariant.
Proof
Let A be a set in , then we need to show that m*(A + y) = m*A ! y.
Let {I
n
} be a countable collection of open intervals that cover A, then m*A = inf
A·I
n
lI
n
.
Clearly {I
n
+ y} will cover A + y, with m*(A + y) = inf
A·I
n
lI
n
+ y, but l(A
n
+ y) = l(A
n
), so
inf
A·I
n
lI
n
= inf
A·I
n
lI
n
+ y = m*(A + y) = m*A. QED.
Problem 8
Prove that if m*A = 0, then m*(A B) = m*B.
Proof
If A · B then A B = B which is trivially true. So we may assume A 1 B. Let {I
n
} denote open
covers of A, and {J
n
} denote open covers of B. Thus {I
n
} {J
n
} will cover A B. Furthermore
m*(A B) s inf
AB· I
n
J
n
lI
n
+ lJ
n
= inf
AB· I
n
J
n
lI
n
+ lJ
n
=
inf
A·I
n
lI
n
+ inf
B·J
n
lJ
n
= m*A + m*B = m*B.
But also B · A B so m*B s m*(A B). Thus we must have that m*(A B) = m*B. QED.
MATH6301.nb 4
Real Analysis  Math 630
Homework Set #2  Chapter 3
by Bobby Rohde
91400
Problem 9
Show that if E is a measurable set, then each translate E + y of E is also
measurable.
Proof
E is measurable so A, we have m*A = m*(A E) + m*(A E
). We also know from Problem 7
that outer measure is translation invariant.
Thus m*(A + y) = m*A = m*(A E) + m*(A E
) = m*((A E) + y) + m*((A E
) + y), y.
m*((A E) + y) + m*((A E
) + y) = m*((A + y) (E + y)) + m*((A + y) (E
+ y)), from the
nature of intersection. But any set B may be written as A + y, thus B, m*B = m*(B (E + y)) +
m*(B (E
+ y)) = m*(B (E + y)) + m*(B ~(E + y)), since ~(E + y) must equal (E
+ y) from
the fact that both ~ and + are 11 operations on sets.
The last equality thus shows that E + y is measurable, QED.
Problem 10
Show that if E
1
and E
2
are measurable, then m(E
1
E
2
) + m(E
1
E
2
) =
mE
1
+ mE
2
Proof
Since each is measurable we know that mA = m(A E
1
) + m(A E
1
), mA = m(A E
2
) + m(A
E
2
), ! A so mE
2
= m(E
2
E
1
) + m(E
2
E
1
) and mE
1
= m(E
1
E
2
) + m(E
1
E
2
)
Thus mE
1
+ mE
2
= 2 * m(E
2
E
1
) + m(E
2
E
1
) + m(E
1
E
2
).
However A B = (A B
) (A
B) (A B).
On inspection we see that the three groupings on the right hand side of the last expression are
pairwise disjoint and hence m(A B) = m(A B
) + m(A
B) + m(A B). Thus 2 * m(E
2
E
1
)
+ m(E
2
E
1
) + m(E
1
E
2
) = m(E
2
E
1
) + m(E
2
E
1
). . m(E
1
E
2
) + m(E
1
E
2
) = mE
1
+
mE
2
, QED.
Problem 11
Show that the condition mE
1
< is neccesary in Proposition 14 by giving a
decreasing sequence <E
n
> of measurable sets with = E
n
and mE
n
=
n.
Example
Let E
n
= (n, o), ! n e . Thus mE
n
= o ! n e , E
n+1
· E
n
, E
n
is measurable and E
n
= O.
Problem 14
a) Show that the Cantor ternary set has measure 0.
Proof
Let C
0
= [0, 1] and let C
n
denote the C
n÷1
with open intervals of length 3
÷n
removed from the
center of each of the 2
n÷1
continuous interval in C
n÷1
. Thus C
n
is the nth stage in the construction
of the Cantor set ÷ and · C
n
, ! n. Since C
n÷1
· C
n
and mC
0
= 1 < o, by Proposition 14, we
have that m(
n=0
o
C
n
) = lim
n÷o
m C
n
. However ·
n=0
o
C
n
since · C
n
, ! n = 0 s m() s
m(
n=0
o
C
n
) = lim
n÷o
m C
n
= lim
n÷o
2
3
n
= 0. Thus m() = 0. QED
MATH6302.nb 2
b) Let F be a subset of [0, 1] constructed in the same manner as the Cantor
ternary set except that each of the intervals removed at the nth step has
length 3
n
with 0 < < 1. Then F is a closed set, F
dense in [0, 1] and mF
= 1  .
Proof
Let F
n
denote the construction at the nth step.
Closed
Consider F
n
which must be open ! n. Since F
= F
n
, we know that F
must be open since infinite
unions of open sets are open. Thus F
= F must be closed since its the complement of an open set.
QED
F
is dense in [0, 1]
Suppose not, then ¬ a, b e F
[0, 1], with a < b, such that (a, b) F
= O. Thus (a, b) · F, with a,
b e F. But this contradicts the contruction of F since every continuous interval must have its
middle removed for some F
n
and (a, b) never did since the entire interval is in F. Thus ! a, b e F
[0, 1], with a < b, ¬ c e (a, b) F
. Hense F
is dense in [0, 1]. QED
mF = 1 
At each phase 2
n÷1
intervals are removed of size o3
÷n
. Hence mF = m([0, 1]) 
mIntervals Removed = 1 
n=1
o 2
n÷1
+o
3
n
= 1 
o
2
n=1
o
2
3
n
= 1 
o
2
*
2
3
1÷
2
3
= 1  o. . mF = 1  o.
QED
Problem 15
Show that if E is measurable and E P, then mE = 0.
Proof
Suppose BWOC that mE > 0. Then we know that mE = m(E +
y), ! y. Let <E
i
> be a sequence of
disjoint sets defined by E
i
= E +
r
i
, with < r
i
>
n=1
o
being an enumeration of the rational numbers in
[0, 1). Clearly E
i
· [0, 1] so m( E
i
) s m[0, 1] = 1, but since E
i
are pairwise disjoint, m( E
i
) =
mE
i
= o. Which gives a contradiction. Hence mE = 0. QED
MATH6302.nb 3
Problem 16
Show that, if A is any set with m*A > 0 then there is a nonmeasurable set E
A.
Proof
Let A
n
denote A [n, n + 1), ! n e
Since A
n
are disjoint we know that m*A = m*( A
n
) = m*A
n
= ¬ n such that m*A
n
> 0. Let P
denote the standard nonmeasurable set, <r
i
> be an enumeration of the rationals and P
i
= P +
r
i
.
Let B = (A
n
 n). Let B
i
= B P
i
. By problem 15, if B
i
is measurable then it must must have
measure 0. Since {B
i
} are pairwise disjoint and B
i
= B we must have that 0 < m*B s
i=1
o
m*B
i
=
0. Which is a contradiction thus ¬ a set · B which is not measurable. Finally by translation
invariance of m* we know that ¬ a set · A
n
· A which must be not measurable.
MATH6302.nb 4
Real Analysis  Math 630
Homework Set #3  Chapter 3
by Bobby Rohde
92100
Problem 18
Show that (v) does not imply (iv) in Propostion 18 by constructing a
function f such that {x : f(x) > 0} = E, a given nonmeasurable set, and such
that f assumes each value at most once.
CounterExample
Let E = the standard nonmeasurable set on [0, 1]
Let f(x) =
x, x E
x, x E
defined on domain [0, )
Thus {x : f(x) > 0} = E, and f(x) = has at most one solution for any , namely or .
Hence {x : f(x) = } is measurable but {x : f(x) > 0} is not measurable.
Problem 19
Let D be a dense set of real numbers. Let f be an extended realvalued
function on such that {x : f(x) > } is measurable D. Show that f is
measurable.
Proof
We wish to show that {x : f(x) > o} is measurable ! o e D is equivalent to the condition {x : f(x) >
o} is measurable ! o.
Take 0 e D then define A
n
= {x : f(x) > 5
n
, with 5
n
e (0
1
n
, 0) D}.
We know that (0
1
n
, 0) D is nonempty ! n since, D is dense.
Consider
n=1
o
A
n
·
n=1
o
x : f x > 0 ÷
1
n
= {x : f(x) ± 0}. But {x : f(x) ± 0} ·
n=1
o
A
n
since ! n, 5
n
< 0. Thus
n=1
o
A
n
= {x : f(x) ± 0} which means that {x : f(x) ± 0} is measurable and
hence by Proposition 18, {x : f(x) > 0} is measurable, so {x : f(x) > o} is measurable ! o. QED
Problem 20
Show that the sum and product of two simple functions are simple.
Proof
AB
A
B
\
AB
=
1, if x e AB
0, if x e AB
\
A
+ \
B
=
1, if x e A and x e B
0, if x e A or x e B
But x e A and x e B = x e A B and, xeA or xe B = x e A B, thus \
AB
= \
A
+ \
B
,
QED.
MATH6303.nb 2
AB
A
B
A
B
\
AB
=
1, if x e AB
0, if x e AB
\
A
+ \
B
÷ \
A
+ \
B
= \
A
+ \
B
÷ \
AB
=
1, if x e A or x e B
0, if x e Aand x e B
But x e A or x e B = x e A B and, xeA and xeB = x e A B, thus
\
AB
= \
A
+ \
B
÷ \
A
+ \
B
. QED.
A
1
A
\
A
=
0, if x e A
1, if x e A
1 ÷ \
A
=
1 ÷ 1 = 0, if x e A
1 ÷ 0 = 1, if x e A
Thus \
A
= 1 ÷ \
A
. QED
Problem 21
a) Let D and E be measurable sets and f a function with domain D E.
Show that f is measurable if and only if its restrictions to D and E are
measurable.
Proof
f is measurable = x : f x > o, x e D E is measurable ! o.
A ÷ x : f x > o, x e D E = B ÷ x : f x > o, x e D C ÷ x : f x > o, x e E
Clearly if B and C are measurable then A is measurable as the union of measurable sets.
Since D is measurable, we have that A ~ D is measurable if A is measurable, but A ~ D = C, so C
would be measurable, and similarly A ~ E = B gives B measurable. Thus if A is measurable B and
C are measurable.
Since B and C are equivalent to f restricted to D and E respectively, we must have that f is
measurable if and only if its restrictions to D and E are measurable. QED.
MATH6303.nb 3
b) Let f be a function with measurable domain D. Show that f is
measurable iff the function g defined by g(x) = f(x) for x D and g(x) = 0
for x D is measurable.
Proof
f is measurable = x : f x > o, x e D is measurable ! o.
! o ± 0, {x : g(x) > o} = x : f x > o, x e D, thus {x : g(x) > o} is measurable iff f is
measurable.
! o < 0, x : gx > o = x : f x > o, x e D x : x e D.
However the first term on the right is clearly measurable iff f is measurable, and the second is just x
e D
, but D
is measurable since D is measurable. Thus x : gx > o is measurable for o < 0 iff f is
measurable.
. x : gx > o is measurable ! o iff f is measurable, so g is measurable iff f is measurable. QED
Problem 23
Prove Proposition 23 by establishing the following lemmas:
a) Given a measurable function f on [a, b] that takes the values ± only on
a set of measure 0, and given > 0, M such that f M except on a set of
measure less than / 3.
Proof
We know that A = x : f x ± ÷o and B = x : f x s o are measurable for all o.
Thus C
o
= A B is measurable = x : f x s o.
We also know that C
o
C
o+1
, since o < o + 1.
Furthermore we know that since f is infinite on only a set of measure 0, then ¬ o
0
such that mC
o
0
<
o. Thus by Proposition 14, we have that m
n=1
o
C
o
0
+n
= lim
n÷>o
m C
o
0
+n
= 0. Hence by
definition of limit, ¬ D = C
o
0
+n
, for some n, with mD < e / 3. QED.
MATH6303.nb 4
b) Let f be a measurable function on [a, b]. Given > 0 and M, a simple
function such that f(x)  (x) < except where f(x) M. If m f M,
then we may take so that m M.
Proof
Let c be a simple function with values of n*(e / 2) where n e ((M+1)*2/e, (M+1)*2/e). Thus
simply choose the value for n that satisfies f(x)  c(x) < e when f(x) < M = e  f(x) <  c(x) < f(x)
+ e = e + f(x) > c(x) > f(x)  e = e + f(x) > n*(e / 2) > f(x)  e = 1 + f(x) / e > n / 2 > f(x) / e  1 = 2 +
2 * f(x) / e > n > 2 * f(x) / e  2.
Such an n being guaranteed to exist by the fact that ¬ at least one integer in every range 1 + a > x >
a  1. Thus we have defined c such that it has the neccesary property. This construction will
clearly yield results of the form m s c s M, when m s f s M, except possibly within e / 2 of the
bounds where we may choose the bound itself to satisfy our criterion.
c) Given a simple function on [a, b], a step function g on [a, b] such that
g(x) = (x) except on a set of measure less than / 3. If m M then we
can take g such that m g M.
Proof
Let 5 =
b÷a
2
n
, define [a
m
, b
m
] = [a + (m1)*5, a + m*5 ], with m e [0, 2
n
]. Thus [a
m
, b
m
]
partition [a, b] into n intervals of width 5.
We know that c(x) =
i=1
N
0
i
*\
A
i
, with A
i
= {x : c(x) = 0
i
}.
Define gx = c/
m
, ! x e a
m
, b
m
, with /
m
= 0
i
such that m(A
i
[a
m
, b
m
)) is the max over all i.
Clearly g(x) is a step function and in the limit n ÷ o we have that g(x) = c(x). Also we have that
m({x : g(x) = c(x)}) =
m=1
2
n
m(A
i
[a
m
, b
m
)), for the A
i
associated with /
m
.
Hence K
n
= m({x : g(x) = c(x)}) =
m=1
2
n
m(A
i
[a
m
, b
m
)).
However K
n
forms a strictly decreasing series since for successive n, each pair of new partition
pieces must cover at least as well as the original. And we know that limit K
n
÷ 0 so, ¬ n such that
K
n
< e / 3 ! e > 0. Thus ¬ some step function g conforming to the same bounds as c that will have
g(x) = c(x) except on a set of measure less than e / 3.
MATH6303.nb 5
d) Given a step function g on [a, b], a continuous function h on [a, b] such
that h(x) = g(x) except on a set of measure less than / 3. If m g M then
we can take h such that m h M.
Proof
Clearly for any interval [c, d], we may construct a continuous curve joining g(c) and g(d).
Since g is a step function, ¬ a partition x
0
< x
1
< ... < x
N
such that for each interval (x
n
, x
n+1
), g(x)
assumes only one value. Let v
n
x be any continuous function joining g(x
n
÷
e
6+N+1
) and g(
x
n
+
e
6+N+1
), which does not exceed its boundary points, for n e [1, N1].
Define h(x) =
gx, if x ÷ x
n
>e
6+N+1
, ! n
v
n
x , if x ÷ x
n
se
6+N+1
Thus h(x) is neccesarily a continuous function and h(x) = g(x) except for N intervals of length
2+
e
6+N+1
, which has total measure
e+N
3+N+1
<
e
3
. Thus by explicit construction we have found aa
continuous h satisfying the conditions. QED.
Problem 29
Give an example to show that we must require mE < in Proposition 23.
Example
Let E = (o, o). Let f
n
x =
x
n
. Thus f
n
x ÷ 0.
Take e > 0. Then f
n
x ÷ 0 = f
n
x < e ! n ± some N implies that  x  < N*e.
Hence B = (N*e, N*e) is the collection of all x such that f
n
x < e ! n ± N. Thus B
÷ A = (o,
N*e] [N*e, o) is the smallest set such that ! x e A
and ! n ± N, f
n
x < e. This is true
because if we were to remove any element of A, we would add an element to A
with f
n
x ± e.
Hence the measure of any set, K, with the neccesary property of Proposition 23 will be larger than
mA = o. Thus ! 5 > 0 we have mK = o > 5. Hence showing a counterexample.
MATH6303.nb 6
Problem 30
Prove Ergoroff's Theorem: If <f
n
> is a sequence of measurable functions
that converge to a realvalued function f a.e. on a measurable set E of finite
measure, then given > 0, a subset A E with mA < such that f
n
converges to f uniformly on E ~ A.
Proof
Define A
i
· E for i e with mA
i
< 5
i
= 2
÷i
p and a smallest N
i
such that ! x e A
i
, and all n ± N
i
, 
f
n
x ÷ f x < e
i
=
1
i
. We know that A
i
exists by Proposition 24.
Let A = A
i
. Thus mA s mA
i
< 2
÷i
p = p.
Since A
= ~ A
m
= A
m
= ! x e E ~ A that f
n
converges to f since  f
n
x ÷ f x < e
m
÷ 0.
Thus we have f
n
converging with some N
m
depending only on e
m
, when f
n
(x) restricted to A. So it
demonstrates uniform convergence. QED
MATH6303.nb 7
Real Analysis  Math 630
Homework Set #4  Chapter 4
by Bobby Rohde
92800
Problem 1
a) Show that if
f x
0, x is irrational
1, x is rational
Then
R
a
b
f xx b a and R
a
b
f xx 0
Proof
R
...
a
b
f x dx = inf
a
b
cx dx, ! step functions c(x) ± f(x), ! x.
But each step function is composed of a finite collection of open intervals, and in the domain each
open interal ¬ an rational number since the rationals are dense. Thus c(x) ± f(x), ! x implies that
the value over any open interval of the step function c(x) is ± 1. Thus the infimum of the integral
of all such step functions must be when c(x) = 1, ! x. .
R
...
a
b
f x dx = inf
a
b
cx dx = b ÷ a + 1 = b ÷a.
R
...a
b
f x dx = sup
a
b
cx dx, ! step functions c(x) s f(x), ! x. Since irrationals are dense, each
interval in the composition of c has an irrational in its domain. Thus c(x) s f(x), ! x = that over
any open interval in c, c(x) s 0. Thus the supremum of the integral of all such step functions must
be when c(x) = 0. Hence R
...a
b
f x dx = sup
a
b
cx dx = 0. QED
b) Construct a sequence <f
n
> of nonnegative, Riemann integrable
functions such that f
n
increases monotonically to f. What does this imply
about changing the order of integration and the limiting process.
Construction
Let f
n
= x
1
n
over (0,1], so limit n we have f = 1.
0
1
f
n
x =
n1
n
. So
lim
n
0
1
f
n
x 1
0
1
f x
0
1
lim
n
f
n
x.
Thus we have that the limit of the integral is the integral of the limit.
MATH6304.nb 2
Problem 2
a) Let f be a bounded function on [a, b], and let h be the upper envelope of f.
Then R
a
b
f
a
b
h.
Proof
R
...
a
b
f x dx = inf
a
b
cx dx, ! step functions c(x) ± f(x), ! x.
We know from Problem 2.51(a) that h(x) ± f(x), and from 2.51(b) we have that h(x) is upper semi
continuous so that h(y) ± lim
x÷y
...........
hx = inf
5>0
sup
0<x÷y<5
hx.
From 2.51(a), hx = f x if and only if f is upper semicontinuous at x, so c(x) ± h(x), except
possibly where f(x) is not upper semicontinuous = f(y) < lim
x÷y
...........
f x. However
hy = inf
5>0
sup
x÷y<5
f x so h(y) = lim
x÷y
...........
f x, when f(y) < lim
x÷y
...........
f x. Let A = {x : c(x) < h(x)}
Suppose A = O. Let , e A. So h(,) > c(,), and h(,) = lim
x÷,
...........
f x. Let e = h(,)c(,). By
definition of lim sup we have that ! 5 > 0 such that  sup f(x)  h(,)  < e / 2 and 0 <  x  ,  < 5, ¬ o
in that range such that f o + e 2 ÷ h, < e 2 = ÷ f o ÷ e 2 + h, < e 2 =
÷ f o + h, < e = ÷ f o + h, < h, ÷c, = f o > c,. Which means that o and ,
must be in differant intervals of the partition of c, however this works ! 5 > 0 which implies that ,
must be a boundary point between intervals of the partition. However since c is a step function it is
composed of a finite number of distinct intervals and thus has a finite number of boundary points.
So A has at most a finite number of items in it.
Thus c ± h except on a set of measure 0, so R
...
a
b
f x dx = inf
a
b
cx dx ±
a
b
hx dx.
Define <c
n
x> to be a sequence of step function such that c
n
is over a equipartition of [a, b] into 2
n
pieces, with the value of c
n
over each piece to be sup h(x) over that piece.
BWOC Suppose lim
n÷o
c
n
x = h(x). That implies that ¬ e > 0 such that ! n,  c
n
x  h(x)  ± e =
! partition intervals including x, ¬ , in that interval with h(,) ± h(x) + e. However h(x) ±
lim
/÷x
...........
h/, which gurantees that for some small distance c > 0, h(x) + e > h(/) ! / e (xc, x+c).
Thus we have a contradiction since h(,) must be < h(x) + e for , within c of x.
Thus lim
n÷o
c
n
x = h(x) and clearly c
n
is bounded since h and f are bounded. Thus by
MATH6304.nb 3
Proposition 6 we have that
a
b
h dx = lim
n÷o
a
b
c
n
dx. Furthermore c
n
are each step functions ± f
so that lim
n÷o
a
b
c
n
dx ± inf
a
b
c dx = R
...
a
b
f x dx, where c is any step function ± f.
Hence
a
b
h dx ± R
...
a
b
f x dx. So using the previous bound we have R
...
a
b
f =
a
b
h. QED.
b) Use Part (a) to prove that: A bounded function f on [a, b] is Riemann
integrable if and only if the set of points at which f is discontinuous has
measure 0.
Proof
A bounded function f is Riemann integrable if and only if R
...
a
b
f = R
...a
b
f .
Since f is bounded we know R
...
a
b
f =
a
b
h, where h is the upper envelope from (a). By symmetry
of defintion, interchanging the supremums and infimums we know that R
...a
b
f =
a
b
g, where g is
the lower envelope.
So f is Riemann integrable if and only if
a
b
h =
a
b
g. However 2.51(c) gives us that g(x) = h(x) if
and only f is continuous at x. So by Proposition 5(i), we know that
a
b
h =
a
b
g =
a
b
h ÷ g = 0.
But since hx ± gx, ! x we also know that
a
b
h ÷ g = 0 if and only if, the set of points were
hx = gx has measure 0, since otherwise ¬ some o > 0 such that m(x : hx ÷ gx = o) > 0,
and hence
a
b
h ÷ g ± o*(measure of that set) > 0. But the set of points hx = gx is identical to
the set of the discontinuities in f. . f is Riemann integrable if and only if the set of points at which f
is discontinuous has measure 0. QED.
MATH6304.nb 4
Prove parts (iii)  (v) of Proposition 5
(iii) If f g a.e., then
E
f
E
g
Hence
E
f
E
f
Proof
E
f s
E
g =
E
f ÷
E
g s 0 =
E
f ÷ g s 0. However f s g a.e. = f ÷ g s 0 a.e.
Thus for simple functions c s f ÷ g s 0 a.e., sup
E
c s 0. Since c s 0 except on a set of measure
0. However we know that the integral must exist since the integral of f and g exist. . since it exists
it must equal sup
E
c, so
E
f ÷ g s 0 =
E
f s
E
g. QED.
It follows immediately that
E
f ±
E
f , since  f  ± f. But also  f  =  f  ± f, so
E
f ±
E
÷ f
= ÷
E
f s
E
f , = ÷
E
f s
E
f s
E
f =
E
f s
E
f . QED
(iv) If A f(x) B, then
A mE
E
f B mE
Proof
Since A s f(x) s B, we know from (iii) that
E
A s
E
f s
E
B. However the integral of a constant
is just that constant times the measure of the set integrated over, so
E
A = A+ mE and
E
B = B+ mE. Hence A+ m E s
E
f s B+ mE. QED.
(v) If A and B are disjoint measurable sets of finte measure, then
AB
f
A
f
B
f
Proof
AB
f =
f + \
AB
=
f + \
A
+ \
B
, since \
AB
= \
A
+ \
B
.
However from (i) we have that
f + \
A
+ \
B
=
f + \
A
+
f + \
B
=
A
f +
B
f
MATH6304.nb 5
Real Analysis  Math 630
Homework Set #5  Chapter 4
by Bobby Rohde
100500
Problem 3
Let f be a nonnegative measurable function. Show that
f 0 implies
that f 0 a.e.
Proof
BWOC, suppose f = 0 a.e. Then ¬ some set E with mE > 0 such that f x > 0, ! x e E.
Since x f x > 0 =
ne
x f x ±
1
n
, we know that the countable sum of the measures of the
sets on the righthand side is greater than or equal to the measure of the left hand side which is non
zero. So ¬ N such that if A = x f x ±
1
N
, then mA > 0.
Define hx = min f x,
1
N
, ! x. Thus hx s f x, ! x. Let F c A, such that 0 < mF < o.
We know
f = sup
g s f
g, and
f ±
F
f , the latter coming from the fact that since f is
nonnegative and
F
f =
f + \
F
with f + \
F
s f , ! x. This implies
f ±
F
f = sup
g s f
F
g ±
F
h ±
1
N
+ m F > 0. Thus
f > 0, which contradicts the
assumption that
f = 0, which implies that f = 0 a.e. QED
Problem 4
Let f be a nonnegative measurable function.
a) Show that an increasing sequence
n
of nonnegative simple functions each of
which vanishes outside a set of finite measure such that f lim
n
.
Construction
Define c
n
=
m÷1
2
n
, ! x e ÷n, n such that
m÷1
2
n
s f x <
m
2
n
and ! me with m s 2
2 n
0, otherwise
Thus c
n
assumes exactly 2
2 n
1 nonzero values over a set of measure s 2n. Thus c
n
is a
nonnegative simple function and vanishes except on a set of finite measure.
Furthermore we can show that c
n+1
± c
n
! x and n as follows. Suppose c
n
x = 0 then clearly
c
n+1
x ± 0, thus suppose c
n
x =
m÷1
2
n
for some integer m with 0 < m s 2
2 n
. This implies that
m÷1
2
n
s f x <
m
2
n
=
2 m÷1
2
n+1
=
m÷1
2
n
s f x <
m
2
n
=
2 m
2
n+1
, but if m÷ 1 s 2
2 n
, then clearly
2 m÷ 1 s 2+ 2
2 n
< 2
2 n+1
, so thus we have that c
n+1
x ±
2 m÷1
2
n+1
=
m÷1
2
n
= c
n
x, so its an
increasing sequence.
Now we need only show that f = lim c
n
.
Let [x] denote the least natural number ± x, and log
2
x = to the logarithim base 2 of x, with log
2
0 =
o.
Given e > 0 and some x
0
, we may choose N = max{[x
0
], ÷log
2
e, [log
2
f x
0
]}.
We wish to show that ! n > N, c
n
x
0
÷ f x
0
< e.
Since n > x
0
± x
0
, we have that x
0
e ÷n, n so x
0
meets the first condition for assigning a
nonzero value to c
n
x
0
.
Since n > [log
2
f x
0
] > log
2
f x
0
, we know that 2
n
> 2
log
2
f x
0
= f x
0
. So
f x
0
<
m
2
n
s
2
2n
2
n
= 2
n
, for some m e , m s 2
2 n
. Thus x
0
meets the second condition for
assigning a nonzero value to c
n
x
0
.
Finally since n > ÷log
2
e] and n meets the requirements to assign x
0
a particular nonzero value,
we know that c
n
x ÷ f x s to the spacing of the values of c
n
=
1
2
n
<
1
2
÷log
2
e
s
1
e
÷1
= e. Thus
c
n
x ÷ f x < e. . c
n
converges pointwise to f ! x. So f = lim c
n
.
MATH6305.nb 2
Thus we have constructed a c
n
satisfying all required properties, QED.
b) Show that
f sup
over all simple functions f.
Proof
By definition
f = sup
h s f
h, where h is a bounded measurable function with m x : hx = 0 is
finite.
We may of course limit ourselves to taking the sup over all h' x such that h' x = max hx, 0.
However all such h' x are nonnegative measurable functions. The measurability of h' x deriving
from the fact that x : hx ± 0 and the set x : hx < 0 must both be measurable.
Hence we may apply part a) and conclude that ! h' x, ¬ an increasing sequence < c
n
> such that
c
n
÷ h' x. Thus by Proposition 10 we have that
h' = lim
c
n
which equals the sup
c
n
since it
is an increasing sequence. Thus the sup over some subset of all such simple functions c must be =
sup
h s f
h =
f . So
f s sup
c.
However c s f, and since we are considering a sup we may assume that c ± 0. Thus by Proposition
8 we have that
c s
f . So
f ± sup
c.
.
f = sup
c, over all simple functions c s f. QED
MATH6305.nb 3
Problem 5
Let f be a nonnegative integrable function. Show that the function F
defined by
Fx
x
f
is continuous by using Theorem 10.
Proof
In order to show continuity we need that Fa = lim
x÷a
Fx, ! x. = We need
÷o
a
f = lim
x÷a
÷o
x
f .
Define f
n
=
f x, x < a ÷
1
n
0, otherwise
then f
n
is an increasing sequence with f
n
÷ f as n÷o ! x e (o,
a).
! x < a, we have that
÷o
x
f <
÷o
a
f
N
for some N such that x < a ÷
1
N
< a = N >
1
a÷x
.
Thus lim
x÷a
÷
÷o
x
f s lim
n÷o
÷o
a
f
n
, and by Theorem 10 we have that lim
n÷o
÷o
a
f
n
=
÷o
a
f . But
also we know that ! x such that 0 < a  x < 5 < 1, we have that
÷o
x
f >
÷o
a
f
M
, ! M <
1
5
, thus as 5
÷ 0 in the limit x ÷ a
÷
, we get that lim
x÷a
÷
÷o
x
f ± lim
n÷o
÷o
a
f
n
.
So lim
x÷a
÷
÷o
x
f = lim
n÷o
÷o
a
f
n
=
÷o
a
f .
Now consider lim
x÷a
+
÷o
x
f =
÷o
a
f + lim
x÷a
+
a
x
f , by Proposition 12. We know that since f is
integrable f < o a.e. Since if f = o on some set, A, of measure > 0, we would have
f ± mA+ o = o, which is false since
f < o, by definition of integrable. Thus without loss
of generality we may assume that f < o everywhere since this won't change it's integral. Now since
f is finite over a closed interval we know that f is bounded over that interval. Let N(x) be the
maximum of f y ! y e [a, x]. Hence,
a
x
f s m[a, x]*N(x), by Proposition 5. However it is clear
that Nx
1
s Nx
2
, if a < x
1
s x
2
. Thus lim
x÷a
+
a
x
f s lim
x÷a
+
ma, x + Nx = 0.
Thus lim
x÷a
+
÷o
x
f =
÷o
a
f + lim
x÷a
+
a
x
f =
÷o
a
f . Hence
÷o
a
f = lim
x÷a
÷o
x
f = Fa =
lim
x÷a
Fx. . F is continuous.
MATH6305.nb 4
Problem 6
Let f
n
be a sequence of nonnegative measurable functions that
converge to f, and suppose f
n
f n. Then
f lim
f
n
Proof
By Fatou's Lemma we know that
f s lim
......
f
n
. Also since f
n
s f ! n, we know that
f
n
s
f ,
! n, and thus lim
......
f
n
s
f .
However lim
......
f
n
s lim
......
f
n
by definition of lim sup and lim inf. So by squeezing,
lim
......
f
n
=
f = lim
......
f
n
= lim
f
n
. QED
Problem 7
a) Show that we may have strict inequality in Fatou's Lemma.
Proof
Consider the sequence < f
n
> defined by f
n
x = 1 if n s x < n + 1, with f
n
x = 0 otherwise.
! x, ¬ N > x, such that f
n
x = 0, ! n > N. Thus f
n
÷ f = 0. So
f = 0, however lim
......
f
n
= 1,
since ! n, we have
f
n
= 0 +
n
n+1
1 = 1. Thus 0 < 1 =
f < lim
......
f
n
. QED
b) Show that the Monotone Convergence Theorem need not hold for
decreasing sequences of functions.
Proof
Let f
n
x = 0 if x < n and f
n
x = 1 for x ± n.
! x, ¬ N > x, such that f
n
x = 0, ! n > N. Thus f
n
÷ f = 0. Hence
f = 0. However consider
lim
f
n
.
f
n
±
n
n+1
f
n
= 1. Hence lim
f
n
± 1 = 0. Hence lim
f
n
=
f . QED
MATH6305.nb 5
Real Analysis  Math 630
Homework Set #6  Chapter 4
by Bobby Rohde
101200
Problem 10
a) Show that if f is integrable over E then so is f and
E
f
E
f
Does the integrability of f imply integrability of f?
Proof
f is integrable f
and f
are each integrable. However f = f
+ f
, so by Proposition 15ii,
we have that f is integrable and
E
f
E
f
E
f
. Also
E
f
E
f
E
f
E
f
E
f
, but since f
and f
are nonnegative we can drop the
asbsolute value signs, so
E
f
E
f
E
f
E
f . So we have shown the first part.
f f
f
, So f integrable ( f
+ f
) is integrable. However the set A x f
x 0
is disjoint from the set B x f
x 0, and A and B must be measurable iff f is a measurable
function. Thus by Proposition 15,
E
f
f
AB
f
f
A
f
B
f
. Thus f
and
f
are each integrable iff f is a measurable function.
But this implies that
A
f
B
f
AB
f
f
E
f , is integrable. Thus f implies
integrability of f iff f is a measurable function.
b) The improper Riemann integral of a function may exist without the
function being integrable. If f is integrable, show that the improper
Riemann integral is equal to the Lebesgue integral whenever the former
exists.
Proof
Let R
a
b
f dx, be a Riemann integrable function with improper limit point a, and f defined for all x
e (a, b]. Note that if any other point in (a, b] is improper, we may into two or more integrals each
of which with one improper limit point so we need only assume one limit is improper. In particular
we know from analysis that improper points must be disconnected in order for a function to be
Riemann integrable. WLOG I assume that a < b, but all parts of the proof may be reversed for a > b,
and will still hold accordingly.
Case a <
Then consider the sequence of measurable integrable functions < f
n
>, such that
f
n
x =
f x, a + 1 n s x s b
0, otherwise
. Then lim f
n
x = f x a.e. Also we have that f
n
s f . Thus by
Theorem 16 we have that
a
b
f = lim
a
b
f
n
. But by construction we must have that f
n
x is
bounded, which allows us to apply Proposition 4 to get that
a
b
f
n
= R
a
b
f
n
, and hence
lim
a
b
f
n
= limR
a
b
f
n
= R
a
b
lim f
n
= R
a
b
f . So
a
b
f = R
a
b
f .
Case a = 
Then consider the sequence of measurable integrable functions < f
n
>, such that
f
n
x =
f x, ÷n s x s b
0, otherwise
. Then lim f
n
x = f x a.e. Also we have that f
n
s f . Thus by
Theorem 16 we have that
a
b
f = lim
a
b
f
n
. But by construction we must have that f
n
x is
bounded, which allows us to apply Proposition 4 to get that
a
b
f
n
= R
a
b
f
n
, and hence
lim
a
b
f
n
= limR
a
b
f
n
= R
a
b
lim f
n
= R
a
b
f . So
a
b
f = R
a
b
f . QED
MATH6306.nb 2
Problem 11
If ø is a simple function, we have two definitions for
ø,
ø =
ø
+

ø

and
ø =
i=1
n
a
i
mA
i
. Show that they are equal.
Proof
Consider
c =
c
+
÷
c
÷
, By problem 4b, we know that ! nonnegative measurable functions f,
f = sup
c, with the sup taken over all simple functions s f, and the right hand side evaluated
according to the old rule for integrating step functions. Since it follows immediately from the
definition of f
+
and f
÷
that c
+
and c
÷
are simple, we can thus deduce that
c
+
= sup
c =
i=1
n
1
b
i
mB
i
,
c
÷
= sup
c =
i=1
n
2
c
i
mC
i
, where c
+
=
i=1
n
1
b
i
\
B
i
and
c
÷
=
i=1
n
2
c
i
\
C
i
. So
c =
c
+
÷
c
÷
=
i=1
n
1
b
i
mB
i
÷
i=1
n
2
c
i
mC
i
, also we know that
B
i
C
j
= O ! i, j and each b
i
corresponds to some a
j
> 0 and each c
i
corresponds to some
a
j
< 0, so
i=1
n
1
b
i
mB
i
÷
i=1
n
2
c
i
mC
i
=
i=1
n
1
b
i
mB
i
+
i=1
n
2
÷c
i
mC
i
=
i=1
n
a
i
mA
i
. Since the
sets C
i
and B
i
, must follow the same correspondence. QED.
Problem 12
Let g be an integrable function on a set E and suppose that < f
n
> is a
sequence of measurable functions such that f
n
x < gx a.e. on E. Then
E
lim
......
f
n
< lim
......
E
f
n
< lim
......
E
f
n
<
E
lim
......
f
n
Proof
We know for free that lim
......
E
f
n
s lim
......
E
f
n
, since this is true for any lim sups and infs.
We only need to show that
E
lim
......
f
n
s lim
......
E
f
n
and lim
......
E
f
n
s
E
lim
......
f
n
.
f
n
is measurable and bounded by an integrable function, therefore f
n
is integrable. Also we know
that lim
......
f
n
s f
n
! n thus By Proposition 15iii, we have that
E
lim
......
f
n
s
E
f
n
! n =
E
lim
......
f
n
s lim
......
E
f
n
. A similar argument for lim sup f
n
± f
n
gives us that lim
......
E
f
n
s
E
lim
......
f
n
,
and thus we are done. QED
MATH6306.nb 3
Problem 15
Let f be integrable over E. Then given > 0,
a) a simple function such that
E
f
Construction
Since 1
E
 f ÷ c  ± 1
E
f ÷ c  = 1
E
f ÷ 1
E
c  = 1
E
f
+
÷ 1
E
f
÷
÷1
E
c , it suffices to show that the
latter is less than e. If we further specify that A = ¦x  f
+
(x) = 0¦ and B = ¦x  f
÷
(x) = 0¦, then
we may clearly choose c = 0, ! x e E  (A  B). So we may further simplify the above
expression to 1
E
f
+
÷ 1
E
f
÷
÷1
E
c  = 1
E
f
+
÷1
E
f
÷
÷ 1
A
c ÷ 1
B
c  ± 1
E
f
+
÷ 1
A
c  ÷1
E
f
÷
+1
B
c ,
which again all that we need to show is that this new expression is less than e.
By Problem 4, we know that ! nonnegative measurable functions f we have that
1
f = sup
1
c,
taken over all simple functions c s f. Since f
+
and f
÷
are nonnegative measurable functions we
know there must exist simple functions c
S
and c
I
respectively such that 0 s
1
f
+
÷
1
c
S
< e l 2
and 0 s
1
f
÷
÷
1
c
I
< e l 2. The second implies that 0 s
1
f
÷
+
1
(÷c
I
) < e l 2. Thus if we
choose c over A = c
S
and c over B = ÷c
I
, then we have that
1
E
f
+
÷ 1
A
c  ÷1
E
f
÷
+1
B
c  < e l 2 + e l 2 = e. Thus we have shown the construction as
required.
a) a step function such that
E
f
Construction
By the Question 3.23c, which part of the proof to Proposition 22, we know that ¬ a step function c
such that c = a simple function c except on a set of less than e / 3 and if M bounds c then M bounds
c, ! e > 0. Let e > 0 and c be a step function such that 1
E
 f ÷ c  < e l 2. Let N be a number such
that c(x)  < N, ! x. This is possible since simple functions are always bounded. Choose c such
that c = c except on a set of measure
e
4+N
with the set denoted by A, and c bounded by N. Thus
1
E
 f ÷c  = 1
EA
 f ÷c  + 1
A
 f ÷c  < 1
EA
 f ÷ c  + 1
A
 f ÷ c  = 1
EA
 f ÷c  +
1
A
 f ÷c +c ÷c  s 1
EA
 f ÷ c  + 1
A
( f ÷ c  + c ÷ c ) s 1
EA
 f ÷c  + 1
A
 f ÷c  + 1
A
c ÷c 
s e / 2 + 2*N *mA = e / 2 + 2*N*
e
4+N
= e.
Therefore 1
E
 f ÷c  < e.
MATH6306.nb 4
a) a continuous function g such that
E
f g
Construction
By question 3.23d, we have that a continuous function g such that g = a step function except on
a set of measure less than / 3, which agress with the bounds on . With this we may replace the
occurances of in the above proof with and with g and the proof will precede identically,
except that we have to infer the boundedness of from the boundedness of the preceding . Thus
we may construct g such that
E
f g .
MATH6306.nb 5
Real Analysis  Math 630
Homework Set #7  Chapter 5
by Bobby Rohde
101900
Problem 1
Let f be the function defined by f 0 0 and f x x sin
1
x
for x 0.
Find D
f 0, D
f 0, D
f 0, and D
f 0.
D
f 0
D
f 0 = lim
h0
f hf 0
h
= lim
h0
hsin
1
h
h
= lim
h0
sin
1
h
= 1, since for arbitrarily small positive h we
can find a value of
1
h
which gives sine its maximal value.
D
f 0
D
f 0 = lim
h0
f hf 0
h
= lim
h0
hsin
1
h
h
= lim
h0
sin
1
h
= 1, since for arbitrarily small positive h we
can find a value of
1
h
which gives sine its minimal value.
D
f 0
D
f 0 = lim
h0
f 0f h
h
= lim
h0
hsin
1
h
h
= lim
h0
sin
1
h
= 1, since for arbitrarily small positive h
we can find a value of
1
h
which gives sine its maximal value.
D
f 0
D
f 0 = lim
h0
f 0f h
h
= lim
h0
hsin
1
h
h
= lim
h0
sin
1
h
= 1, since for arbitrarily small positive h
we can find a value of
1
h
which gives sine its minimal value.
Problem 2
a) Show that D
f x = D
f x
Proof
D
+
[÷ f (x)] = lim
h÷0
+
.......
÷f (x+h)÷[÷f (x)]
h
= lim
h÷0
+
.......
÷
f (x+h)÷f (x)
h
= inf
h÷0
+
sup
0<esh
÷
f (x+e)÷f (x)
e
] =
inf
h÷0
+
÷ inf
0<esh
f (x+e)÷f (x)
e
 = ÷

'
.
.. sup
h÷0
+
inf
0<esh
f (x+e)÷f (x)
e
\
!
.
.. = ÷ lim
.......
h÷0
+
f (x+h)÷f (x)
h
= ÷D
+
f (x). QED.
b) If gx = f x, then D
gx = D
f x
Proof
÷D
÷
f (÷x) = ÷ lim
.......
h÷0
+
f (÷x)÷f (÷x÷h)
h
=÷ lim
.......
h÷0
+
g(x)÷g(x+h)
h
= lim
.......
h÷0
+
÷
g(x)÷g(x+h)
h
= lim
.......
h÷0
+
g(x +h) ÷
g(x)
h
=
D
+
g(x). QED
MATH6307.nb 2
Problem 3
a) If f is continuous on [a, b] and assummes a local maximum at c (a, b),
then
D
f c D
f c 0 D
f c D
f c
Proof
Well D
+
f c = lim
h÷0
+
.......
f c+h÷f c
h
± D
+
f c = lim
.......
h÷0
+
f c+h÷f c
h
, since lim sups are neccesarily greater
than lim infs. Similarly D
÷
f c s D
÷
f c. So we only really need to establish that
D
+
f c s 0 s D
÷
f c. Since f c is a local maximum ¬ an open interval I · a, b such that c
e I and f c is the maximum of f over I. WLOG we may consider h to be sufficiently small such
that c + h and c ÷h e I.
Consider D
+
f c = lim
h÷0
+
.......
f c+h÷f c
h
, since c + h e I, f c +h < f c = f c + h ÷ f c < 0.
Furthermore since h > 0, we know that
f c+h÷f c
h
< 0. By definition lim
h÷0
+
.......
gh = inf
h>0
sup
0</<h
g/,
but as h ÷ 0 , the supremums form a nonincreasing sequence, and hence so long as we permit ± o
as acceptable limits, lim
h÷0
+
.......
gh must exist.
Thus we have that lim
h÷0
+
.......
f c+h÷f c
h
s 0, since each
f c+h÷f c
h
< 0, and the limit exists.
Similarly D
÷
f c = lim
.......
h÷0
+
f c÷f c÷h
h
has
f c÷f c÷h
h
> 0 over I, and the limit must exist over the
extended reals, so lim
.......
h÷0
+
f c÷f c÷h
h
± 0. Hence D
+
f c s 0 s D
÷
f c. QED.
b) What if f has a local maximum at a or b?
Answer
If f is maximal at a or b then the two derivates which are still inside the interval must exist and obey
the above inequality, however the other two may not even exist. Thus with a < b and f a a local
maximum we have that D
+
f a s D
+
f a s 0, and if f b is a local max then
0 s D
÷
f b s D
÷
f b.
MATH6307.nb 3
Problem 4
Prove: If f is continuous on a, b and one of its derivates (say D
+
) is
everywhere nonnegative on a, b, then f is nondecreasing on a, b; i.e.
f x < f y for x < y.
Proof
« Lemma  For a function gx such that D
+
gx > e > 0, the above property holds.
Suppose ¬ x < y, such that gx > gy. Since g is continuous we may apply the normal knowledge
of continuous functions to conclude that either g has a local maximum value or g is always
decreasing. If g were always decreasing then we know that D
+
gx = lim
h÷0
+
.......
gx+h÷gx
h
< 0 since
gx + h ÷ gx < 0, ! x. Which is a contradiction, thus g has a local max.
However by problem 3, D
+
gc s 0, where c is the location of the local max, but D
+
gx > 0, ! x
which gives a contradiction. Thus we know that g is nondecreasing.
« Consider f x = gx  e = x, where gx is defined as in the lemma.
Therefore D
+
f x = lim
h÷0
+
.......
f x+h÷f x
h
= lim
h÷0
+
.......
gx+h÷e+x+h÷gx÷e+x
h
= lim
h÷0
+
.......
gx+h÷gx
h
÷
e+h
h
=
lim
h÷0
+
.......
gx+h÷gx
h
÷ e = lim
h÷0
+
.......
gx+h÷gx
h
÷ e ± 0.
So if f is an arbitrary function of the type given in the problem then we know that ! e > 0,
f x +e + x, is a nondecreasing function. Suppose ¬ x < y such that f x > f y, yet we know that
f x +e + x s f y +e + y. So this means that f x ÷ f y > 0 and f x ÷ f y s e + y ÷ x. Let o =
f x ÷ f y, then o > 0, and o s e +y ÷ x. However we may choose e =
o
y÷x+2
> 0. Which give
that o s e*(yx) =
o
2
. Which is a contradiction. Hence f x must be nondecreasing.
MATH6307.nb 4
Now all we need to show is that D
f x 0 iff the other derivates are 0.
However we know that nondecreasing is equivalent to D
f x 0, and it follows immediately
that f x h f x and f x f x h are each 0 for h > 0. However this means that the terms
for which we are considering the limit of in each derivate must be always 0. So by definition all
the derivates are 0 if D
f x 0.
Now we need only look at the other direction. Clearly for both D
f x D
f x and D
f x
D
f x, so if respectively D
f x and D
f x are 0 than the other two will be. So in order to
complete the other direction we only need that D
f x 0 implies that the function is
nondecreasing. However if we replace D
f x with D
f x and the related definitions in the
above two steps its easy to see that the remainder of the proof will preceed verbatim. Hence
D
f x 0 implies that f x is nondecreasing. This completes the proof. QED.
MATH6307.nb 5
Real Analysis  Math 630
Homework Set #8  Chapter 5
by Bobby Rohde
110300
Problem 7
a) Let f be of bounded variation (BV) on a, b. Show that for each
c e a, b the limit of f x exists as x  c

and also as x  c
+
. Prove that
a monotone function (and hence a function of BV) can have only a
countable number of discountinuities.
Proof
« Lemma: For a monotone BV function g on a, b, the number of discontinuies such that
7 e > 0, gx +e  gx  e
>1
·········
n
, 7 n e , is finite.
BWOC Suppose this number is infinite. Since it is a BV, we know that g is strictly real valued.
However we can start at x = a, and proceed towards b such that each discontinuity we choose e
succifiently small that ¬ infintely many disconinuties between the current location and b. Thus we
experience arbitrarily many increases of at least
1
n
, which since g is monotonic implies that
gb = o, which is a contradiction. Hence the number is finite.
« The number of discontinuities of g as above is at most countable.
Let A
n
= x gx + e ÷ gx ÷ e
>1
n
, ! e > 0}. It is clear that any discontinuity of a monotone
function must be in some A
n
for n sufficiently large. So the set of all discontinuities is
n=1
o
A
n
,
which must be countable since it is a countable union of finite sets. Thus the number of
discontinuities is countable.
The limit of f x exists as x c
and x c
Consider a fixed c. Suppose ¬ e > 0 such that f over c ÷ e, c, contains no discontinuties, then f is
continuous in that region and limit of f x as x ÷ c
÷
exists.
Thus we may assume that ! e > 0 ¬ a discontuinity in f over the region in question. We need to
show that given , > 0, ¬ 5 > 0, such that ! x with 0 < c ÷ x < 5 = f x ÷ f c < ,. However we
know that f is BV and thus the sum of two monotone functions g, h. Thus by the first lemma there
exist only finitely many points with gx ÷ gc
>1
n
. So we may take 5 sufficiently small such that
all points of differance >
,
2
are more than 5 away. Similarly for h. Since they contribute with
opposite sign, f x ÷ f c < ,, were 5 is the minimum of the two 5's needed.
Hence limit x ÷ c
÷
exists. The proof however clearly proceeds likewise for x ÷ c
+
. QED
b) Construct a monotone function on a, b which is discontinuous at each
rational point.
Construction
Let the sequence < r
i
>
i=1
o
, be an enumeration of the rationals in the interval 0, 1. Define
f
r
x =
r, x ± r
0, x < r
.
Let f x =
i=1
o
f
r
i
x +
1
2
n
. Claim that f is such a function as required.
Clearly
i=1
o
f
r
i
x +
1
2
n
s
i=1
o 1
2
n
= 1. Thus since the lefthand side is a monotonically increasing
bounded series it must converge to some number less than infinity. Furthermore it should be clear
that ! x, y e 0, 1 with x < y, we have that ¬ some q e x, y , since rationals are dense. So
f y ÷ f x ± f
q
y +
1
2
N
, for some N since all terms for which f
r
i
x = 0 yields f
r
i
x = f
r
i
y, and
f
q
x = 0 = f
q
y. Thus f x is monotonicly increasing.
Furthermore ! p e , and ! h > 0. f p +h ÷ f p ÷ h ± f
p
p + h +
1
2
N
, using the above property.
However since p e implies that f
p
p + h = f
p
p = p. Thus ! h > 0,
f p +h ÷ f p ÷ h ±
p
2
N
> 0. Which implies that f p +h ÷ f p ÷h ±
p
2
N
, so limit h ÷ 0,
lim
x÷p
+
f x ÷lim
x÷p
÷
f x ±
p
2
N
. Hence f x is discontinuous at p since the limits can not be
equal.
Thus f satisfies all the conditions desired. QED.
MATH6308.nb 2
Problem 8
a) Show that if a c b, then T
a
b
T
a
c
T
c
b
and that hence T
a
c
T
a
b
.
Proof
T
a
b
T
a
c
T
c
b
By definition T
a
b
= sup t, where sup is taken over all possible partitions of a, b. However
t =
i=1
k
f x
i
÷ f x
i÷1
. Then either c = x
i
, for some i, or c e x
i
, x
i+1
, for some i.
If c = x
i
, then clearly we may split the partition over a, b into two partitions, one over a, c and
the other over c, b, such that t
a
b
is equal to t
a
c
+ t
c
b
, for those partitions.
If c e x
i
, x
i+1
then consider the partition a = x
0
< x
1
<... < x
i
< c < x
i+1
<... < x
k
= b. In this
partition t
2
is ± to the original t ÷ t
1
, since f x
i+1
÷ f x
i
= f x
i+1
÷ f c + f c ÷ f x
i
s
f x
i+1
÷ f c + f c ÷ f x
i
, by triangle inequality. If we break the new partition as above we
have two partitions about a, c and c, b, respectively with t taken over these partitions summed =
t
2
± t
1
. Hence T
a
c
+T
c
b
± T
a
b
.
T
a
b
T
a
c
T
c
b
Every partition p
1
, p
2
over a, c and c, b respectively can be used to form a partition over a, b
by joining the two at c. However T
a
b
= sup t over all possible partitions, so this includes the
partition formed by the union of p
1
and p
2
and hence T
a
b
T
a
c
T
c
b
.
Thus T
a
b
= T
a
c
+T
c
b
. Showing that T
a
c
s T
a
b
.
b) Show that T
a
b
f g T
a
b
f T
a
b
g, and T
a
b
c f c T
a
b
f .
Proof
T
a
b
f g T
a
b
f T
a
b
g
T
a
b
f + g = sup t f + g, where t f + g =
i=1
k
f + g x
i
÷ f + g x
i÷1
=
i=1
k
f x
i
+ gx
i
÷ f x
i÷1
÷ gx
i÷1
s
i=1
k
f x
i
÷ f x
i÷1
+gx
i
÷ gx
i÷1
by the triangle
inequality, however this last term is just t f + tg.
Thus sup t f + g s sup t f +tg, hence T
a
b
f + g s T
a
b
f + T
a
b
g. QED
MATH6308.nb 3
T
a
b
c f c T
a
b
f
T
a
b
c+ f = sup tc+ f , where tc+ f =
i=1
k
c+ f x
i
÷ c+ f x
i÷1
=
i=1
k
c + f x
i
÷ f x
i÷1
= c +
i=1
k
f x
i
÷ f x
i÷1
= c + t f .
Thus sup tc+ f = sup c + t f = c + sup t f , hence T
a
b
c+ f = c + T
a
b
f . QED
Problem 10
a) Let f be defined by f 0 0 and f x x
2
sin
1
x
2
, for x 0. Is f of BV
on 1, 1?
Answer
No. Over any interval a, b, the diffence from x
2
sin
1
x
2
, will be = a
2
sin
1
a
2
÷b
2
sin
1
b
2
.
Suppose a is a minimum, thus
1
a
2
= n+ r ÷
r
2
, for some integer n. Thus ¬ an adjacent maximum at
b such that
1
b
2
= n+ r +
r
2
. So a
2
sin
1
a
2
÷b
2
sin
1
b
2
=
1
n+r÷
r
2
÷1 ÷
1
n+r+
r
2
+ 1=
÷2
2+n÷1+r
+
÷2
2+n+1+r
=
÷8+n
4 n
2
÷1+r
. However as the number of partitions gets arbitrarily large we may
add arbitrarily many terms of this form with n increasing. So
T
÷1
0
f ±
n=1
o
÷8+n
4 n
2
÷1+r
±
n=1
o
2+
1
n+r
= o, since the last series is a constant times the
harmonic series. Therefore f is not a BV.
MATH6308.nb 4
b) Let g be defined by g0 0 and gx x
2
sin
1
x
, for x 0. Is g of BV
on 1, 1?
Answer
Yes. Since g(x)  is symmetric in x we know that it is suffiecent to show that g(x) is BV on [÷1, 0].
Consider a single transition for a minimum to an adjacent maximum. It is clear that it will proceed
monotonically in this region. On inspection it is clear that t over a monotonic region will be
independant of the choice of partition and have the value equal to the differance of the end points.
(Either p or n is 0).
From problem 8a) and an obvious induction it is clear that T
a
b
may be broken into finitely many
pieces. We know that minima and maxima occur at sin(
1
x
) = ± 1 = x =
1
n+r±
r
2
. It is obvious that
T
÷1
÷2
r
<o, purely by inspection and our knowledge of basic calculus. Proceed by breaking T
a
b
into m
pieces such that the first piece is T
÷1
÷2
r
, and the proceeding m2 pieces follow for decreasing values
of n, and the last piece represents the tail going to 0.
Now we need to evaluate T over one these pieces that have been set up to run from a local max to a
local min or vice versa. So the points in question have the form 
1
n+r+
r
2
,
1
(n÷1)+r+
r
2
. So

1
n+r+
r
2
]
2
+ sin(n+r +
r
2
) ÷
1
(n÷1)+r+
r
2
]
2
+sin((n ÷1) + r +
r
2
)  = 
1
n+r+
r
2
]
2
+ 
1
(n÷1)+r+
r
2
]
2
 =

2+(n+r)
2
+2+(
r
2
)
2
(n+r+
r
2
)
2
((n÷1)+r+
r
2
)
2
 < 
2+(n
2
+
1
4
)
(n÷1)
2
+(n÷2)
2
 < 
2+(n
2
+
1
4
)
(n÷2)
4
, note that the inequalities depend on the fact
that n < 0.
Now we want to consider the total T
÷1
0
s T
÷1
÷2
r
+_
n=÷1
÷(m÷2)

2+(n
2
+
1
4
)
(n÷2)
4
 + T
tail
0
, where T
tail
0
is the part not
covered in m steps. Taking limit m ÷ o, the tail contribution must go to 0. So we only need that
the _
n=÷1
÷o

2+(n
2
+
1
4
)
(n÷2)
4
 < o. However this may be rewritten as _
n=1
o

2+(n
2
+
1
4
)
(2÷n)
4
. However this is
bounded above by
3
n
2
for n large, which is a convergent sequence. Thus our sum must converge.
Hence T
÷1
0
< o. So T
÷1
1
< o = x
2
sin(
1
x
) is BV. QED.
MATH6308.nb 5
Problem 14
a) Show that the sum and differance of two AC functions are also AC.
Proof
Let f and g be AC. Let e > 0. Then ¬ 5
1
, 5
2
such that
i=1
n
f x
i
·
÷ f x
i
<e
2
and
i=1
n
gx
i
·
÷ gx
i
<e
2
! finite collection of intervals x
i
, x
i
·
, such that
i=1
n
x
i
·
÷ x
i
< 5 = min 5
1
, 5
2
.
However
i=1
n
f x
i
·
+ gx
i
·
÷ f x
i
÷ gx
i
s
i=1
n
f x
i
·
÷ f x
i
+ gx
i
·
÷ gx
i
, by triangle
inequlaity. Further
i=1
n
f x
i
·
÷ f x
i
+ gx
i
·
÷ gx
i
<
e
2
+
e
2
= e.
Hence we may choose 5 = min 5
1
, 5
2
in order to satisfy the condition and show that f + g is AC.
Since f x
i
·
÷ f x
i
= ÷ f x
i
·
÷ f x
i
= ÷ f x
i
·
÷ ÷ f x
i
, we know that f is AC implies
that f is AC. So f ÷ g must also be AC.
b) Show that the product of two AC functions is AC
Proof
Lemma: f is AC implies that f
2
is AC.
Let f be AC with respect to a, b. Then we know that f is BV, and hence that since a, b is a
closed interval, that f is bounded. Let M > 0 be a bound on f . Let e > 0. Since f is AC, ¬ 5 >0
such that
i=1
n
f x
i
·
÷ f x
i
<e
2 M
, ! finite collection of intervals x
i
, x
i
·
, such that
i=1
n
x
i
·
÷ x
i
< 5.
Consider
i=1
n
f x
i
·
2
÷ f x
i
2
=
i=1
n
f x
i
·
÷ f x
i
+ f x
i
·
+ f x
i
s
i=1
n
f x
i
·
÷ f x
i
+ f x
i
·
+ f x
i
s
i=1
n
f x
i
·
÷ f x
i
+ 2 M <
e
2 M
+ 2 M = e.
Thus f
2
is AC. Concluding the Lemma.
MATH6308.nb 6
If f , g are AC over a, b then f g is AC.
f + g =
f +g
2
÷ f
2
÷g
2
2
, but that both sums, differances and squares of AC functions are AC, so we
only need that constant multiples of AC functions are AC. But for c+ f , where f is AC it follows
immediately that for any e > 0 we may choose f constrained by
e
c
, and the c will carry through to
make c+ f , AC. (Note c = 0 is obviously true.)
Hence f + g can be gotten as a sequence of operations that perserve AC, so it is also AC. QED.
c) If f is AC on a, b and if f is never zero there, then the function g
1
f
is
also AC on a, b.
Proof
Since f is never 0 over a closed interval, we know that ¬ m > 0 such that f > m everywhere. Let e
> 0. Since f is AC, ¬ 5 > 0, such that
i=1
n
f x
i
·
÷ f x
i
< e + m
2
.
Consider
i=1
n
1
f x
i
·
÷
1
f x
i
=
i=1
n
f x
i
÷f x
i
·
f x
i
·
+ f x
i
s
i=1
n
f x
i
÷f x
i
·
m
2
<
e+m
2
m
2
= e.
Thus g =
1
f
is AC. QED.
MATH6308.nb 7
Real Analysis  Math 630
Homework Set #10  Chapter 6
by Bobby Rohde
111700
Problem 7
a) For 1 < p < o, we denote l
p
the space of all sequences < t
v
>
v=1
o
such
that
v=1
o
t
v
p
< o. Prove the Minkowski ineqality for sequences
< t
v
+ n
v
>
p
< < t
v
>
p
+ < n
v
>
p
.
Here we have 1 < p < o,
< t
v
>
p
p
=
v=1
o
t
v
p
and < n
v
>
o
= sup n
p
Proof
« Case 1 < p < o
a < /
v
> a
p
= (_
v=1
o
(/
v
)
p
)
1
p
= a < /
v
+ p
v
> a
p
= (_
v=1
o
( /
v
+p
v
)
p
))
1
p
.
If either /
v
or p
v
has norm 0, the proof is trivial so we may assume that each has norm greater than
1 and then define real numbers o, 0 > 0 such that a <
/
v
o
> a
p
= 1, a <
p
v
0
> a
p
= 1.
(_
v=1
o
( /
v
+ p
v
)
p
))
1
p
s (_
v=1
o
( /
v
 +p
v
)
p
))
1
p
= _
v=1
o
o 
/
v
o
 +0 
p
v
0
]
p
]]
1
p
=
_
v=1
o
(o + 0)
p

o
o+0
] 
/
v
o
 +
0
o+0
] 
p
v
0
]
p
]]
1
p
s _
v=1
o
(o + 0)
p

o
o+0
] 
/
v
o
]
p
+ 
0
o+0
] 
p
v
0
]
p
]]
1
p
,
since p ± 1 and the quantities inside the distribution are each less than 1.
However expnading this expression and summing we see that it is in turn = (o + 0) + o, with o s 1.
So (_
v=1
o
( /
v
+ p
v
)
p
))
1
p
s o + 0 = a < /
v
> a
p
+ a < p
v
> a
p
. QED.
« Case p = o
< p
v
+ /
v
>
o
= sup /
v
+p
p
s sup /
v
+ p
v
s sup /
v
+ sup p
v
=
< /
v
>
o
+ < p
v
>
o
. QED.
b) Show that if < t
v
> e l
p
and < n
v
> e l
q
with
1
······
p
+
1
·····
q
= 1, then
v=1
o
t
v
= n
v
< < t
v
>
p
= < n
v
>
q
Proof
« Case p = 1, q = o
v=1
o
/
v
+p
v
s
v=1
o
/
v
+sup p
v
=
v=1
o
/
v
+ < p
v
>
o
= < /
v
>
1
+ < p
v
>
o
.
« Case 1 < p < o
Define o
v
= p
v
q
p
. We may chosoe to assume that /
v
and p
v
are positive since the norm is dependant
only on their absolute value. By Lemma 3 = p+ t + /
v
+ p
v
= p+ t + /
v
+ o
v
p÷1
s o
v
+ t + /
v
p
÷ o
v
p
Summing both sides.
v=1
o
p+t + /
v
+ p
v
s
v=1
o
o
v
+ t + /
v
p
÷o
v
p
=
< o
v
+ t + /
v
>
p
p
 < o
v
>
p
p
s < o
v
>
p
+ < t + /
v
>
p
p
÷ < o
v
>
p
p
Differentiating with respect to t we have that
v=1
o
p+ /
v
+ p
v
s p+ < /
v
>
p
+ < o
v
>
p
p÷1
=
< /
v
>
p
+ < p
v
>
q
. QED.
Problem 10
Let < f
n
> be a sequence of functions in L
o
. Prove that < f
n
> converges
to f in L
o
if and only if there is a set E of measure 0 such that f
n
converges
uniformly to f on E
.
.
Proof
f
n
÷ f in L
o
imples f
n
÷ f
o
÷ 0. Thus ! e ÷ 0 ¬ N such that ! n > N, f
n
÷ f
o
< e. =
except on a set E of measure 0, f
n
÷ f s e. Thus f
n
is uniformly converges to f on E
.
If f
n
uniformly converges to f except on a set E of measure 0 then ! e > 0, ¬ N such that ! x e E
and n > N, f
n
x ÷ f x < e. = except on E, f x < f
n
x + e s ess sup f
n
x + e. Thus f is
bounded a.e. and hence f is in L
o
.
MATH63010.nb 2
Problem 11
Prove that L
is complete.
Proof
In order to show this we need that every Cauchy sequence of functions < f
n
> in L
o
converges to
some f in L
o
.
Given e > 0, ¬ N such that ! m, n > N, f
n
÷ f
m
< e, by definition of Cauchy. Since this is L
o
norm we know that f
n
÷ f
m
is bounded a.e. by e.
Define the pointwise limit f x =
lim
n÷o
f
n
x, if the limit exists
o, otherwise
.
It suffices to show that ess sup f < o, since if f
n
converges then it must converge to f a.e.
Since f
n
÷ f
m
is bounded a.e. by e, except on a set M of measure 0. However if we consider the
collection of all such M ! m > n, we have a countable collection of sets of measure zero so there
union is also a set of measure 0. Thus f
n
x ÷ f x < e if lim
n÷o
f
n
x exists. Since f
n
is in L
o
it
is bounded by some ess sup o. Then where the limit exists, f is bounded by o + e.
Thus we only need to show that the limit must exist upto a set of measure 0.
MATH63010.nb 3
Problem 16
Let < f
n
> be a sequence of functions in L
p
, 1 < p < o, which converge a.e.
to a function f in L
p
. Show that < f
n
> converges to f in L
p
if and only if
f
n
 f .
Proof
If a f
n
a ÷ a f a = a f
n
a ÷ a f a ÷0 =
(
1
( f
n
)
p
)
1
p
÷(
1
( f )
p
))
1
p
 ± (
1
( f
n
)
p
÷( f )
p
))
1
p
 ÷0, since p ± 1. However  f
n

p
÷( f )
p
 ±
( f
n
 ÷  f )
p
, for p ± 1
< f
n
> converges to f in L
p
= a f
n
÷ f a÷ 0 = (
1
( f
n
÷ f )
p
)
1
p
÷ 0. However, ( f
n
÷ f )
p
±
( f
n
 ÷ f )
p
.
??????????????????????????????????
Demonstrate Proposition 8 fails at p = o.
Show that ¬ f , g e L
o
, such that 7 step functions ¢ and continuous
functions ¢, f  ¢
o
> e > 0 and g  ¢
o
> e > 0.
Counterexample on step functions.
Define f (x) = 
sin(
1
x
), x = 0
0, x = 0
.
Clearly f (x) is bounded by 1 and thus in L
o
. Consider a step function c on partition
A = ¦0 = /
0
< /
1
< /
2
<... < /
N
= 1¦. Let 5 denote the size of the smallest interval in the partition.
Consider intervals of the form A
n
= 
1
n+r
,
1
(n+1)+r
, for n e . Clearly A
n
· [0, 1], and ¬ infinitely
many A
n
of length < 5. So choose n, m such that A
n
· [/
m
, /
m+1
] and length A
n
< 5. Clearly by
construction  f (A
n
)  = [0, 1], so regardless of the value of c over [/
m
, /
m+1
],  f ÷ c 
±1
2
over A
n
.
Hence a f ÷ c a
o
±
1
2
. Thus showing the counter example to Proposition 8.
MATH63010.nb 4
Counterexample on continuous functions.
Define gx =
0, x s
1
2
1, x >
1
2
.
Clearly g is in L
o
. Consider a continuous function c approximating g. By continuity at x =
1
2
, ! e
> 0 ¬ 5 > 0 such that ! x ÷
1
2
< 5, cx ÷c
1
2
< e.
Suppose that c
1
2
= 0, then choose e =
1
2
and ¬ the interval
1
2
,
1
2
+5 where cx <
1
2
, yet g = 1
over this interval so g ÷ c
o
>
1
2
.
Suppose that c
1
2
= 0, then choose e =
c
1
2
2
and ¬ the interval
1
2
÷ 5,
1
2
where cx >
c
1
2
2
.
So g ÷ c
o
±
c
1
2
2
. Thus g ÷c
o
> 0. Thus showing the counter example for Proposition
8 at p = o. QED.
MATH63010.nb 5
Real Analysis  Math 630
Homework Set #11  Chapter 6 & 11
by Bobby Rohde
112500
Problem 6.21
a) Let g be an integrable function on 0, 1. Show that a bounded
measurable function f such that f 0 and
f g g
1
f
Proof
Case g
1
0
Let f x = sgngx, where sgno =
o
o
, o = 0
0, o = 0
. Since g
1
= 0, that implies that
m x : gx = 0 > 0 = ess sup f(x) = 1.
Hence
f + g =
g + sgng =
g = g
1
= g
1
+ f
o
. QED
Case g
1
0
Let f x = o > 0, ! x. Then f
o
= o.
f + g = o+
g s o+
g = 0 ! finite o. Thus
f + g = g
1
+ f
o
= 0. QED.
b) Let g be a bounded measurable function. Show that 7 e > 0 ¬ an
integrable function f such that
f = g > g
o
 e = f
1
Proof
By definition ess sup g(x) = inf M : m t : f t > M = 0. Given e > 0. Let
E = x : gx > g
o
÷ e. Thus mE > 0. Let f x = \
E
. Then
f + g =
\
E
+ g ±
\
E
+ g
o
÷ e by definition of E. Hence we have this = mE+ g
o
÷ e =
g
o
÷ e + f
1
. Thus
f + g ± g
o
÷ e + f
1
. QED.
Problem 6.22
Find a representation for the bounded linear functionals on l
p
, 1 < p < o.
Construction
Problem 6.24
Show that the element g in L
p
given by Theorem 13 is unique.
Proof
MATH63011.nb 2
Problem 11.7
Prove Proposition 4: If X, , is a measure space, then we can find a
complete measure space X,
0
,
0
such that
i.
0
.
ii. E E
0
E
iii. E
0
E A Bwhere B and A C, C , C 0.
Proof
First we must show that the
0
defined by (iii) is in fact a oalgebra. Thus we need that !
E
1
, E
2
e
0
, = E
1
E
2
e
0
and E
1
e
0
and the union of any countable collection of sets in
0
is in
0
.
Clearly E
1
e
0
= E
1
= A
1
B
1
where B
1
e and A
1
· C
1
, C
1
e , µC
1
= 0. And similarly for
E
2
. Thus E
1
E
2
= A
1
A
2
B
1
B
2
. And we know that B
1
B
2
e , since is a o
algebra. Furthermore A
1
A
2
· C
1
C
2
. C
1
C
2
e , since is a oalgebra and
0 s µC
1
C
2
s µC
1
+ µC
2
= 0. So µC
1
C
2
= 0.
E
1
= A
1
B
1
= A
1
B
1
= C
1
A
1
B
1
B
1
C
1
. However B
1
C
1
e , and
hence B
1
C
1
e . Also C
1
A
1
B
1
· C
1
. Hence E
1
has the appropriate form to be in
0
.
Thus we need only show that the union of a countable collection of sets in
0
is in
0
. Consider
n=1
o
E
n
, E
n
e
0
= A
n
B
n
= B
n
A
n
, we know that B
n
e , so we only need
that A
n
is the subset of some set of measure 0. However A
n
· C
n
, with µC
n
= 0, and the union
of countable collection of things of meaurse 0 is still measure 0. Hence it suffices to consider the
union of the C
n
and we are done. Thus
0
is a oalgebra.
Consider E e
0
with E = AB as above so that A is a subset of a set of measure 0. We wish to
show that µB is not dependant on the choice of B e . Given two such representations
A
1
B
1
= A
2
B
2
. Let A
1
· C
1
, with µC
1
= 0 and A
2
· C
2
, with µC
2
= 0. Thus,
B
1
A
1
A
2
B
2
=µB
1
± µC
2
+ µB
2
= µB
2
, however by symmetry of argument, we have
µB
2
± µC
1
+ µB
1
= µB
1
, and hence µB
1
= µB
2
.
Thus we may define µ
0
E = µB, without concern that the value will change depending on our
choice of B. So that µ
0
A = 0 if A · C, µC = 0.
MATH63011.nb 3
It then follows immediately that µ
0
is a measure from the fact that µ is a measure and that we have
only added sets which have measure 0 and the sets generated by their unions with sets of , which
will neccesarily preserve the additivity of the measure.
Thus we have created a complete measure with the neccesary properties.
Problem 11.10
Prove Proposition 7: Let f be a nonnegative measurable function . Then
a sequence
n
of simple functions with
n1
n
such that f lim
n
at each point of X. If f is defined on a finite measurable space, then we
may choose the fucntions
n
so that each vanishes outside a set of finite
measure.
Construction
Define E
n,k
= x : k + 2
÷n
s f x < k + 1 + 2
÷n
and c
n
= 2
÷n
k=0
2
2+n
k + \
E
n,k
, with respect to each
pair of integers < n, k >.
I wish to show that c
n
thus defined has the neccesary properties.
Since f is finite, for a given < n, k >, n > 0, k > 0, mE
n,k
< o. Thus m
k=1
2
2+n
E
n,k
< o, and so c
n
=
0, except on that set which is finite. So each function vanishes off a set of finite measure.
Furthermore it should be clear that for a given n, the sets E
n,k
are mutually exclusive and that if for
a given n and k, f x is bracketed by k
0
+2
÷n
, k
0
+ 1 + 2
÷n
then for the next n, we will have f x
bracketed by either 2+ k
0
+ 2
÷n+1
, 2+ k
0
+1 + 2
÷n+1
or
2+ k
0
+1 + 2
÷n+1
, 2+ k
0
+ 2 + 2
÷n+1
. And the lower bound of the interval is the same as the
value returned by the sum, it is then clear that c
n+1
± c
n
.
Furthermore, since the width of the interval between k + 2
÷n
and k + 1 +2
÷n
goes to 0 as n ÷ 0,
while the range goes to both 0 and o as k ÷0 and k ÷ 2
2+n
. This allows that for any value of f x,
it will eventually be in some interval and the value can never exceed f x. Thus since the interval
successively approximates f x with an e width equal to that of the interval hence the limit must go
to f x.
MATH63011.nb 4
Problem 11.17
Prove Proposition 15: If f and g are integrable functions and E is a
measurable set, then
i.
E
c
1
f c
2
g c
1
E
f c
2
E
g
ii. If h f and h is measurable then h is integrable.
iii. If f g a.e., then
f
g.
Proof
E
c
1
f c
2
g c
1
E
f c
2
E
g
Simply break f and g into + and  parts and apply Proposition 13.
If h f and h is measurable then h is integrable.
If we can first show part (iii), then by applying that result we can conclude that
÷o < ÷ f s
h s f < o, and thus h in integrable. So let us proceed to part (iii).
If f g a.e., then
f
g.
If we consider
f = inf
csf
c, then clearly ¬ a sequence of monotone increasing simple functions
< c
n
> which converge to g which will be less than f a.e. By montone convergence theorem
g = lim
c
n
s inf
csf
c =
f . Hence
f ±
g. QED.
MATH63011.nb 5
Real Analysis  Math 630
Homework Set #12  Chapter 12
by Bobby Rohde
112500
Problem 12.2
Assume that E
i
is a sequence of disjoint measurable sets and E = E
i
.
Then A we have that
A E
A E
i
Proof
If < E
i
> has only one nonempty set element then this is trivial. Suppose < E
i
> has a finite
number of nonempty sets, then by the assumption of the induction hypothesis say that
E
=
i=1
n÷1
E
i
, conforms to the property that µ
+
AE
=
i=1
n÷1
µ
+
AE
i
.
Consider E =
i=1
n
E
i
, then µ
+
AE = µ
+
AE E
n
+ µ
+
AE E
n
, since E
n
is
measurable. However E E
n
, and exploitning the fact that each E
i
is disjoint, we arrive at
µ
+
AE = µ
+
AE
n
+ µ
+
AE
=
i=1
n
µ
+
AE
i
. Thus the induction is proved. Hence we
know for any finite sequence this property will hold.
So ! N, µ
+
A
i=1
N
E
i
=
i=1
N
µ
+
AE
i
. However A
i=1
N
E
i
· A
i=1
o
E
i
, so
µ
+
AE = µ
+
A
i=1
o
E
i
±
i=1
N
µ
+
AE
i
. The left hand side is now independant of N so
we make take the right hand side to infinity. So µ
+
AE ±
i=1
o
µ
+
AE
i
, yet we know form
the properties of outer measure that µ
+
AE s
i=1
o
µ
+
AE
i
. Hence
µ
+
AE = µ
+
AE
i
. QED.
Problem 12.4
Prove Proposition 9  Let be a semialgebra of sets and a nonnegative set
function defined on with 0if . Then has a unique
extension to a measure on the algebra generated by if the following
conditions are met:
i) If a set C is the union of a finite disjoint collection C
i
of sets in , then
C
C
i
ii) If a set C in is the union of a countable disjoint collection C
i
of sets in , then
C
C
i
a) Condition (i) implies that if A is the union of each of two finite disjoint
collections C
i
and D
i
of sets in , then
C
i
D
j
.
Proof
A C
i
by condition (i), but also by condition (i) we have that A D
j
, so be the
transitivity of equality, C
i
D
j
.
b) Condition (ii) implies that is countably additive on .
Proof
By Problem 5, we have that all sets in take the form of countable unions of sets in . However it
is always possible to generate a disjoint sequence of sets from an arbitrary sequence of sets, which
preserve unions, provide that we may take finite intersections and complements, as we can on a
semialgebra. Thus all elements of may be expressed in terms of the union of a countable dijoint
collection of sets.
This implies that condition (ii) is equivalent to A , A C
i
, for any disjoint sequence
C
i
, such that C
i
A. Which is exactly what is required to show that the measure is
countably additive.
In conclusion, property (i) gives us that the notion of measure on is well defined and unique on
finite sets, and property (ii) gives us that countable additivity is preserved and that the value of A
is still uniquely defined over countably additive unions.
MATH63012.nb 2
Problem 12.5
Let be a semialgebra of sets and the smallest algebra of sets containing
.
a) Show that is comprised of sets of the form A =
i1
n
C
i
with C
i
.
Proof
Clearly it is neccesary that be closed under finite unions and thus it will contain finite unions of
elements in . We need to show that this is sufficient to be closed under finite intersections and
complements.
Let A, B , such that A
i1
n
A
i
, and B
i1
m
B
i
, where each set A
i
and B
i
is in . Then
AB
i1
n
A
i
i1
m
B
i
=
i1
n
A
i
j1
m
B
j
=
i1
n
j1
m
A
i
B
j
, but each A
i
B
j
,
since is closed under finite intersections. Thus AB is a finite union and hence an element of
of the appropriate form. By induction is closed under finite intersections.
Consider A
=
i1
n
C
i
. However by definition of semialgebra, C
i
k1
n
i
D
i,k
, with D
i,k
. Thus
A
i1
n
k1
n
i
D
i,k
, but
k1
n
i
D
i,k
. Thus since is closed under finite intersections from
above, we must have that A
.
Hence is an algebra of sets. That it is the smallest algebra of sets containing is clear from the
fact that any algebra of sets must contain the elements of the form
i1
n
C
i
with C
i
, yet no other
elements have been added besides these.
b) Show that
, so that
and
may be replaced in theorems
by
and
, respectively.
Proof
, thus
. However
contains all elements of the form
i1
C
i
, where C
i
.
Hence for each element in , A =
i1
n
C
i
, A
i1
C
i
. Thus
=
. Thus
=
. QED.
MATH63012.nb 3
Problem 12.6
Let be a collection of sets which is closed under finite unions and finite
intersections; an algebra of sets, for example.
a) Show that
is closed under countable unions and finite intersections.
Proof
Well by definition
o
is the set of all sets generated by countable unions of sets in so all that
remains to be shown is that it is closed under finite intersections. Given
B e
o
with B e , C e . Then we want to show that B C e
o
. We know that there must
exist a sequence of sets < B
i
>, with each set in such that B
i
= B, since it appears in the
closure of with respect to unions.
Define < E
i
> to be the sequence of sets such that E
i
= B
i
C e , since is closed under finite
intersections. However the E
i
= B
i
C = B C. Thus the intersection of any element of
o
with one of is in
o
. By induction we may intersect any finite number of elements in
with elements of
o
.
Given two elements B, C e
o
, B, C e , then we know that B C = B C
i
, where C
i
are
the elements of a generating sequence for C. This is equivalent to B C
i
, however we know
that B C
i
e
o
from above, and thus taking the countable union will also be in
o
. .
o
is
closed under finite intersections.
b) Show that each set in
is the intersection of a decreasing sequence of
sets in
Construction
Clearly if the set in question is a member of
o
then the result is trivial, since we just take every
term in the sequence to be that set.
Given B e
o5
, B e , then ¬ a sequence < B
i
>
i=0
o
of sets in
o
such that B
i
= B.
Define A
i
=
j=i
o
B
j
k=0
i÷1
B
k
, with A
0
= B
i
. Given any element x e B, = x e B
k
, ! k,
hence x e A
i
, ! i. Given y e B, y e B
N
for some N and hence y e A
i
, ! i ± N. If N is the least
such N that works then y e A
i
, ! i < N, and thus A
i
's form a decreasing sequence and A
i
= B.
Thus we have a countable intersection of decreasing sets in A
o
that yields B e A
o5
.
MATH63012.nb 4
Problem 12.7
Let be a finite measure on an algebra , and
the induced outer
measure. Show that a set E is measurable if and only if for each > 0 there
is a set A
, A E, such that
EA .
Proof
MATH63012.nb 5
Real Analysis  Math 630
Homework Set #13  Chapter 12
by Bobby Rohde
121100
Problem 12.19
Let X Y be the set of positive integers, X, and let be
the measure defined by setting E equal to the number of points in E if E
is finite and if E is an infinite set. (This measure is called the counting
measure.) State the Fubini and Tonelli Theorems explicitly for this case.
Answer
While I am unclear on exactly what suffices as response to this question, I am going to say when we
can apply the Fubini and Tonelli theorems and then simpify the conclusions of the theorems as
applicable to this specific case.
Let f x, y be a realvalued function from X ×Y.
First considering the Tonelli theorem, we know that the set of positive integers is ofinite, so we
may apply the Tonelli theorem provided that f is nonnegative everywhere.
In the case of Fubini theorem we need that
X×Y
f dµ×v < o. However since X ×Y is a set of
cardinality b
0
we know that ¬ an enumeration of the points in X ×Y. Let < r
i
>
i=1
o
be such an
enumeration. Then we can equate f x, y =
i=1
o
f r
i
+ \
r
i
x, y. If we further define
c
n
x, y =
i=1
n
f r
i
+ \
r
i
x, y, then the sequence of functions are each simple and clearly
converge to f. Suppose we consider the positive and negative parts of f. Then f
±
± c
n
±
, and by
application of the monotone convergence theorem,
f
±
= lim
n÷o
c
n
±
. However the right hand
side is equivalent to lim
n÷o
i=1
n
f
±
r
i
+ µr
i
=
i=1
o
f
±
r
i
, since the counting measure over a
set of one element is 1. Thus saying that
X×Y
f dµ×v < o is equivalent to asking that
i=1
o
f
+
r
i
< o and
i=1
o
f
÷
r
i
< o.
Let us now consider the implication of the theorems which are of course exactly the same. For
simplicity I will not state the results generate by interchange of x and y.
(i) for almost all x the function f
x
defined by f
x
y f x, y is a measurable function on
Y.
Well f
x
y =
i=1
o
f r
i
+ \
r
i
x, y, where x is fixed. Which is equal to
i=1
o
f x, i + \
i
y.
Consider y : f
x
y ± o = A
o
. We need only show that µA
o
is well defined for all o. However A
o
is some subset of Y, and thus A
o
e Y, so A
o
is measurable, since it is in the collection of
measurable sets, and hence f
x
y is a measurable function by Proposition 11.5.
(ii)
Y
f x, yy is a measurable function of X.
Let us explicitly compute
Y
f x, y dvy =
Y
f
x
y dv =
Y
i=1
o
f x, i + \
i
y dv. Using the
same caveat as above we may consider this as positive and negative pieces of f. Since integration is
a linear operation, we will only consider nonnegative functions f. In the nonnegative case we may
exploit the fact that the sum is the limit of a sequence of increasing simple function and apply the
monotone convergence theorem to say that
Y
f
x
y dv =
i=1
o
f x, i + vi =
i=1
o
f x, i, since vi = 1, ! i .
That the function
i=1
o
f x, i is measurable follows immediately form the fact that x e X, and =
(X). The argument is the same as the 2nd paragraph in (i).
(iii)
X
Y
f x, y
XY
f x, y
Working from the conclusion of (ii), that
Y
f
x
y dv =
i=1
o
f x, i, we will then integrate with
respect to µ, to find that
X
Y
f x, y dv d µ =
i=1
o
j=i
o
f i, j µi. (Note that technically we
msut first consider this as the limit of finite sums, but the functions in the finite sums are increasing
nonnegative functions whose limit is the o sum, and the integrals are thus the same by Monotone
convergence theorem.) However µi = 1 since µ is the counting measure, and thus
X
Y
f x, y dv d µ =
i=1
o
j=i
o
f i, j. However we also know that
X×Y
f x, y dµ×v =
X×Y
i=1
o
f r
i
+ \
r
i
x, y dµ×v =
i=1
o
f r
i
+ µ×v r
i
=
i=1
o
f r
i
, since
µ×v r
i
= 1, ! i . However since the r
i
's range over ever pair j, k of integers exactly once,
this is identical to the above result.
Hence
X
Y
f x, y dv d µ =
X×Y
f x, y dµ×v =
i=1
o
j=i
o
f i, j.
MATH63013.nb 2
Problem 12.22
Let h and g be integrable functions on X and Y, and define
f x, y hxgy. Then f is integrable on XY and
XY
f
X
h
Y
g
Proof
Without loss of generality we may suppose that f , g and h are nonnegative functions. If not we
may rewrite them as the differance of their negative and positive parts and by the linearity of
integration we only need to show the proof for nonnegative functions.
Consider
X×Y
f dµ×v. If the integral exists then by definition of integration
X×Y
f dµ×v =
inf
c±f
X×Y
c dµ×v =
i=1
n
c
i
µ×v E
i
X ×Y, where cx, y =
i=1
n
c
i
\
E
i
x, y is a
nonnegative simple function and each E
i
· X ×Y.
However g, h are integrable functions, so they each admit a representation
X
h d µ = inf
c
h
±h
X
c
h
d µ =
i=1
m
1
d
i
µH
i
X and
Y
g dv = inf
c
g
±g
Y
c
g
dv =
i=1
m
2
f
i
µG
i
Y, with
c
h
x =
i=1
m
1
d
i
\
H
i
x and c
g
y =
i=1
m
2
f
i
\
G
i
y.
Since f x, y = hx + gy = that every c
g
+ c
h
is a possible function for c =
i=1
n
c
i
\
E
i
x, y =
i=1
m
1
d
i
\
H
i
x +
i=1
m
2
f
i
\
G
i
y =
i=1
m
1
j=1
m
2
d
i
f
j
\
H
i
x \
G
j
y. However
\
H
i
x \
G
j
y = \
H
i
×G
j
x, y.
This implies that f admits a representation as a sum of n = m
1
+ m
2
elements, with each
E
i
= H
ji
×G
ki
and c
i
= d
ji
+ f
ki
, where the j, k depend on i and each pair of j, k are used
exactly once over the m
1
+ m
2
terms.
This that
i=1
n
c
i
µ×v E
i
X ×Y =
i=1
m
1
+m
2
d
ji
+ f
ki
µ×v H
ji
×G
ki
X ×Y. We know
that each G
j
and H
k
is in X and Y respectively since these are the basis sets, so this devolves to
i=1
m
1
+m
2
d
ji
+ f
ki
µ×v H
ji
×G
ki
. However H
j
×G
k
is a rectangle in X ×Y, so
µ×v H
ji
×G
ki
= µH
ji
+vG
ki
=
i=1
m
1
+m
2
d
ji
+ f
ki
µ×v H
ji
×G
ki
=
i=1
m
1
+m
2
d
ji
+ f
ki
µH
ji
+ vG
ki
, but by definition of
the ji and ki this is just equal to
i=1
m
1
d
i
µH
i
+
i=1
m
2
f
i
vG
i
=
X
c
h
d µ+
Y
c
g
dv.
MATH63013.nb 3
Thus inf
c±f
X×Y
c dµ×v s inf
c
h
±h
X
c
h
d µ+ inf
c
g
±g
Y
c
g
dv. Evaluating the infs we arrive at
X×Y
f dµ×v s
X
h d µ+
Y
g dv, assuming the integral over f exists. However there is nothing
specific in this argument to the fact we choose inf, except the last inequality, hence by the sup over
simple functions less than the given function approach we can arrive at
X×Y
f dµ×v ±
X
h d µ+
Y
g dv, if the integral over f exists.
However what we have in fact shown is that
X
h d µ+
Y
g dv s sup
csf
X×Y
c dµ×v s inf
c±f
X×Y
c dµ×v s
X
h d µ+
Y
g dv, the middle
inequality is always true of the sups and infs used to define integrals. .
X
h d µ+
Y
g dv = sup
csf
X×Y
c dµ×v = inf
c±f
X×Y
c dµ×v, which implies that
X×Y
f dµ×v
exists and its value is
X
h d µ+
Y
g dv. QED.
Author's note: I am all but sure that there is a better way to do this, but I don't know what it is so I
decided to do it based on first principles.
MATH63013.nb 4
Problem 12.24
The following example shows that we cannot remove the hypothesis that f
be nonnegative from the Tonelli Theorem or that f be integrable from the
Fubini Theorem. Let X Y be the positive integers and be the
counting measure. Let
f x, y
2 2
x
, if x y
2 2
x
, if x y 1
0, otherwise
Contradiction
First it is neccesary to notice that f x, y = 2 ÷2
÷x
\
A
x, y + ÷2 + 2
÷x
\
B
x, y, where
A = x, y : x = y and B = x, y : x = y + 1.
In order to disprove the Fubini and Tonelli thereoms it suffices to show that property (iii) in each
fails to hold. In other words I wish to show that
X
Y
f x, y dv d µ =
Y
X
f x, y d µ dv.
Let us first consider
X
Y
f x, y dv d µ. So computing
Y
f x, y dv at a particular x, we have that
= 2 ÷ 2
÷x
vA
x
+ ÷2 + 2
÷x
vB
x
. Furthermore we know that for a fixed x, A
x
has only one
point determined by x = y, and similarly B
x
has only one point. Hence the counting measure
evaluated on these sets is just 1. Thus
Y
f x, y dv = 2 ÷ 2
÷x
+ ÷2 + 2
÷x
= 0. Hence
X
Y
f x, y dv d µ = 0.
Now let us consider
Y
X
f x, y d µ dv = that we first evaluate
X
f x, y d µ for a fixed y. Thus
gives the integral as 2 ÷ 2
÷y
µA
y
+ ÷2 + 2
÷y÷1
µB
y
, note the substitution in the exponent in
terms of the fixed coordinate. Again each of these has exactly one item in its set, with the
exception of B
y
at y = 1, which is empty because x = 0 is not in the domain of positive integers. So
for y ± 2,
X
f x, y d µ = 2 ÷ 2
÷y
+÷2 +2
÷y÷1
= 2
÷y
1
2
÷ 1 = ÷2
÷y÷1
. Also at y = 1,
X
f x, y d µ = 2 ÷2
÷1
=
3
2
.
Now in order to compute the integral over Y, it is not hard to see that by restricing the integral to
the first N integers we are in fact integrating over a simple function. Thus
Y
X
f x, y d µ dv =
1,2,... ,N
3
2
\
1
+÷2
÷y+1
+ \
1
dv =
3
2
v1 +
i=2
N
÷2
÷i÷1
+ vi Since we are using the
counting measure, each evaluation is in fact one. Simplifying the geometric series we have
MATH63013.nb 5
=3
2
÷
1
2
1÷
1
2
=
1
2
. Thus
Y
X
f x, y d µ dv =
1
2
= 0 =
X
Y
f x, y dv d µ. Hence neither the
Fubini nor the Tonelli theoreoms will hold for this particular fucntion and choice of measure. QED.
Problem 12.25
The following example shows that we cannot remove the hypothesis that f
be integrable from the Fubini Theorem or that µ and v are ofinite from
the Tonelli Theorem: Let X = Y be the interval 0, 1, with = the class
of Borel sets. Let µ be the Lebesgue measure and the v the counting
measure. Then the diagonal A = < x, y > e X ×Y : x = y is measurable
(is an
oö
, in fact), but its characteristic function fails to satisfy any of the
equalities in condition (iii) of the Fubini and Tonelli Theorems.
Contradiction
Again we will seek an exception of the Fubini and Tonelli Theorems by computing the differant
halves of part (iii) in each result and showing that they are not equal.
Consider
X
Y
\
A
x, y dv d µ. First we wish to compute
Y
\
A
x, y dv, for a particular choice of x.
This means that
Y
\
A
x, y dv = vA
x
, but \
A
x
y is only nonzero for the one value where x = y,
and so since v is the counting measure, we have that
Y
\
A
x, y dv = 1. So
X
Y
\
A
x, y dv d µ =
X
1 d µ = µX = 1.
Now consider
Y
X
\
A
x, y d µ dv. First we start with
X
\
A
x, y d µ, for a particular value of y.
Which evaluates the integral to µA
y
, which again is a set containing only the point where x = y.
However since µ is the Lesbegue measure this evaluates to 0. So we have that
Y
X
\
A
x, y d µ dv
=
Y
0 dv = 0. Since 0 = 1, we have thus shown that \
A
fails to satisfy the (iii) property of the
Fubini and Tunelli Theorems. QED.
MATH63013.nb 6
Problem 12.31
Let f be a nonnegative measurable function on o, o, and let m
2
be the
twodimensional Lebesgue measure on
2
. Then
m
2
< x, y > : 0 < y < f x = m
2
< x, y > : 0 < y < f x =
f x dx
Let øt = m x : f x > t. Then ø is a decreasing function and
0
o
øt dt =
f x dx.
Proof
« m
2
< x, y > : 0 < y < f x = m
2
< x, y > : 0 < y < f x =
f x dx
Essentially in the first part we are attempting to show that
f x dx is the same thing as the area
under the curve (with and without its boundary).
Let A = < x, y > : 0 s y s f x and B = < x, y > : 0 < y < f x. Let µ denote the Lesbegue
measure. Consider
\
A
dµ×µ, since is ofinite and \ is everywhere nonnegative, we may
apply Tonelli Theorem to conclude that
\
A
dµ×µ =
\
A
d µ d µ =
µA
x
d µ. For any
particular x we have that A
x
is the interval 0, f x, and thus that µA
x
= f x. Thus
\
A
dµ×µ
=
f x d µ, but the integral with respect to Lesbegue measure is exactly what is intended by the
notation
f x dx. Finally we know that
\
A
dµ×µ = µ×µ A by the definition of an integral
of a characteristic function, but µ×µ = m
2
. Hence we have that m
2
A =
f x dx.
If we then note that B behaves exactly the same as A in the above proof since B
x
= 0, f x, and
hence µB
x
= µ0, f x = f x. Thus the conclusion is equally valid that m
2
B =
f x dx.
MATH63013.nb 7
0
tt
f xx
While in the above proof we made a construction that showed the area was the same as that under
the curve f x, by exploiting a series of vertical bars, in the proof that follows we will attempt to
demonstrate that it is equivalent to a series of horizontal bars.
On the question that c is a decreasing function, this follows immediately from that fact that
mA s mB, whenever A c B, and the observation that if x e x : f x ± t
1
then x e x : f x ± t
2
, !
t
2
s t
1
. Thus for t
1
> t
2
, we have that x : f x ± t
1
. x : f x ± t
2
. So m x : f x ± t
1
±
m x : f x ± t
2
.
Let A = < x, t > : f x ± t ± 0
Consider
0
o
ct dt =
0
o
m x : f x ± t dt =
0
o
mA
t
dt =
0
o
÷o
o
\
A
x, t dx dt. The last
equality coming from the fact that the measure of a crosssection is the same as the integral of its
characteristic function. Since characteristic functions are bounded and both measure are taken with
respect to the ofinite real numbers, we may apply Tonelli thoerem to conclude that
0
o
÷o
o
\
A
x, t dx dt =
÷o
o
0
o
\
A
x, t dt dx. From here we need only observe that A is the same
as A in the first section of this problem, upto the labelling of variables. Thus from above we may
conclude that
÷o
o
0
o
\
A
x, t dt dx =
f x dx. .
0
o
ct dt =
f x dx. QED.
MATH63013.nb 8
MATH6301.nb
2
Problem 3
If there is a set A in Proof
such that mA <
, then m = 0.
= , hence = A, so m(A
By Way of Contradiction (BWOC). Suppose m = 0, and let mA = < . A A and are disjoint. Thus by Property 3, m(A ) = mA + m = + . But A ) = mA = . Thus = + > . Which is a contradiction, hence m = 0. QED.
Problem 4
Let nE be for an infinite set E and equal to the number of elements in E for a finite set. Show that n is a countably additive set function that is translation invariant and defined for all sets of real numbers. Proof
Countably Additive We need to show that disjoint sequences <Ai > of sets in , n( Ai ) = nAi . nAi = , so we may as well assume that Ai is finite i. If Ai is infinite for any i then n( Ai ) = However since Ai is disjoint that implies that the number of elements in Ai A j = nAi + nA j , i j. Thus by repeated applications it is clear that n( Ai ) must equal nAi . QED. Translation Invariant We need to show that n(A + y) = nA y. Clearly the definition of A + y = {x + y : x A}, establishes a 11 correspondance between elements of A and elements of A + y, thus A and A + y must have the same number of elements and hence n(A + y) = nA y. QED Defined on All Sets of Clearly it is in the nature of sets that they must have either an infinite number of elements or exactly one welldefined finite number of elements so it is impossible to construct a set on which does not have a unique value under operation by n. Thus all sets on are measurable by n. QED.
MATH6301.nb
3
Problem 5
Let A be the set of rational numbers between 0 and 1, and let {In} be a finite collection of open intervals covering A. Show that l(In) 1. Proof
Let (an , bn ) denote the interval In with end points an < bn . BWOC Suppose 0 < an , n, then since rationals are dense, a rational number between 0 and min{an }, not covered by In , which is a contradiction. So we know n such that an 0. Similarly we must have an m such that bm 1. an open interval C = BWOC Suppose In with bn < 1 such that m n, am bn < bm is false (bn , min({an : an > bn } {1})), which is not in In . But since the rational numbers are dense, a rational number in C, which contradicts the fact that {In } covers A. Thus In with bn < 1, m such that am bn < bm. Consider T = ( In ) {bn } {an }. Clearly T covers A since In covers A. Furthermore since am bn < bm is true under the conditions just stated, this implies that including {an } and {bn } will close any gaps between intervals. Since there are a finite number of open intervals, we are adding only a finite number of points to T, and hence m*T = l(In ). But T is now a continuous interval with (0, 1) T, using the fact that n, m such that an 0 and bm 1. Thus m*T 1  0 = 1. l( In ) 1. QED.
Problem 6
Prove that: Given any set A and any > 0, an open set O such that A O and m*O m*A + . There is a G G such that A G and m*A = m*G. Proof
Since inf
A In
l In = m*A we know by definition of infimum that
a collection of open intervals In is an open set with m*O =
{In } where l In m*A + > 0. Thus we have that O = l In m*A + . Which gives us the first part of the Proposition
For the second part we merely have to consider G = n 1 O 1 , where O is the O associated with a n specified in the first part. Since A O > 0, we know that A G. Also since G is an
then we need to show that m*(A + y) = m*A y. Thus {In } {Jn } will cover A B. Proof Let A be a set in . m*(A B). we can therefore conclude that m*A = m*G. Thus we must have that m*(A B) = m*B. with 0 in our construction of G. with m*(A + y) = inf A A In inf In l In . QED. QED. G G . A Let {In } be a countable collection of open intervals that cover A. m*A + Problem 7 Prove that m* is translation invariant. Finally since m*A m*O . But also B A B so m*B . Let {In } denote open covers of A. So we may assume A B. and {Jn } denote open covers of B. then m*A = Clearly {In + y} will cover A + y. but l( An y ) = l(An ).nb 4 intersection of a countable collection of open intervals. QED. If A B then A B = B which is trivially true. then m*(A Proof B) = m*B. l In y . Problem 8 Prove that if m*A = 0. Furthermore m*(A B) inf l In l Jn = inf l In l Jn = A B In Jn A B In Jn A inf In l In + inf B Jn l Jn = m*A + m*B = m*B.MATH6301. so inf In l In = inf A In l In y m*(A + y) = m*A.
thus B. we have m*A = m*(A E) + m*(A E). m*((A E) + y) + m*((A E) + y) = m*((A + y) (E + y)) + m*((A + y) (E + y)). Thus m*(A + y) = m*A = m*(A E) + m*(A E) = m*((A E) + y) + m*((A E) + y).Real Analysis . The last equality thus shows that E + y is measurable. y.Math 630 Homework Set #2 . since ~(E + y) must equal (E + y) from the fact that both ~ and + are 11 operations on sets. Proof E is measurable so A. then each translate E + y of E is also measurable. But any set B may be written as A + y. . QED. m*B = m*(B (E + y)) + m*(B (E + y)) = m*(B (E + y)) + m*(B ~(E + y)). from the nature of intersection. We also know from Problem 7 that outer measure is translation invariant.Chapter 3 by Bobby Rohde 91400 Problem 9 Show that if E is a measurable set.
However Cn . m(E1 E2 ) + m(E1 E2 ) = mE1 + mE2 . Example Let En = (n. En 1 En . Thus Cn is the nth stage in the construction of the Cantor set and Cn . we have that m( n 0 Cn ) = limn m C n . Problem 11 Show that the condition mE1 < is neccesary in Proposition 14 by giving a decreasing sequence <En> of measurable sets with = En and mEn = n.nb 2 Problem 10 Show that if E1 and E2 are measurable. n 0 m( ) n 0 Cn since 2 n m( n 0 Cn ) = limn m C n = limn = 0. However A B = (A B) (A B) (A B). Thus m( ) = 0. Thus 2 * m(E2 E1 ) + m(E2 E1 ) + m(E1 E2 ) = m(E2 E1 ) + m(E2 E1 ). Proof Let C0 = [0. QED. Thus mEn = n . then m(E1 mE1 + mE2 Proof E2) + m(E1 E2) = Since each is measurable we know that mA = m(A E1 ) + m(A E1 ). A so mE2 = m(E2 E1 ) + m(E2 E1 ) and mE1 = m(E1 E2 ) + m(E1 E2 ) Thus mE1 + mE2 = 2 * m(E2 E1 ) + m(E2 E1 ) + m(E1 E2 ). by Proposition 14. n . On inspection we see that the three groupings on the right hand side of the last expression are pairwise disjoint and hence m(A B) = m(A B) + m(A B) + m(A B). mA = m(A E2 ) + m(A E2 ). n. ). QED 3 .MATH6302. En is measurable and En = . Since Cn 1 Cn and mC0 = 1 < . Problem 14 a) Show that the Cantor ternary set has measure 0. 1] and let Cn denote the Cn 1 with open intervals of length 3 n removed from the center of each of the 2n 1 continuous interval in Cn 1 .
Suppose BWOC that mE > 0. with ri n 1 being an enumeration of the rational numbers in [0. Thus a. m( Ei ) = mEi = . 1]) * 1 32 = 1 . 1] so m( Ei ) m[0. then a.nb 3 b) Let F be a subset of [0. b F. with a < b. b) F . 1] Suppose not. Thus (a. 1]. b F [0. Proof Let Fn denote the construction at the nth step. QED . Problem 15 Show that if E is measurable and E Proof P. such that (a. Thus F = F must be closed since its the complement of an open set. 1].. Which gives a contradiction. Let <Ei> be a sequence of disjoint sets defined by Ei = E ri . c (a. Clearly Ei [0. we know that F must be open since infinite unions of open sets are open. then mE = 0. Closed Consider Fn which must be open n. 1] = 1. Hense F is dense in [0.. 3 2 mF = 1 . 1).MATH6302.. 1] and mF =1. Since F = Fn . QED F is dense in [0. with a < b. But this contradicts the contruction of F since every continuous interval must have its middle removed for some Fn and (a. y. F dense in [0. Hence mE = 0. QED mF = 1  At each phase 2n QED 1 intervals are removed of size n 1 2n 1 m Intervals Removed = 1  3n =12 2 n n 1 3 3 n. =12 Hence mF = m([0. with a. 1]. b F [0. b) F = . Then F is a closed set. 1] constructed in the same manner as the Cantor ternary set except that each of the intervals removed at the nth step has length 3 n with 0 < < 1. b) F. b) never did since the entire interval is in F. Then we know that mE = m(E y). but since Ei are pairwise disjoint.
. Let P denote the standard nonmeasurable set. n + 1). Which is a contradiction thus a set B which is not measurable. n Since An are disjoint we know that m*A = m*( An ) = m*An n such that m*An > 0. Pi . <ri > be an enumeration of the rationals and Pi = P ri .n). Proof Let An denote A [n. Finally by translation invariance of m* we know that a set An A which must be not measurable. Let Bi = B measure 0.nb 4 Problem 16 Show that. if A is any set with m*A > 0 then there is a nonmeasurable set E A. Since {Bi } are pairwise disjoint and Bi = B we must have that 0 < m*B i 1 m*Bi = 0.MATH6302. if Bi is measurable then it must must have Let B = (An . By problem 15.
.Real Analysis . namely Hence {x : f(x) = } is measurable but {x : f(x) > 0} is not measurable. 1] x. x E Let f(x) = defined on domain [0. ) x. and such that f assumes each value at most once.Chapter 3 by Bobby Rohde 92100 Problem 18 Show that (v) does not imply (iv) in Propostion 18 by constructing a function f such that {x : f(x) > 0} = E. or . and f(x) = has at most one solution for any . CounterExample Let E = the standard nonmeasurable set on [0. x E Thus {x : f(x) > 0} = E.Math 630 Homework Set #3 .. a given nonmeasurable set.
Thus n 1 An = {x : f(x) } which means that {x : f(x) } is measurable and hence by Proposition 18. Let f be an extended realvalued function on such that {x : f(x) > } is measurable D.1 . But {x : f(x) } n 1 x : f x n 1 An n = {x : f(x) since n. QED Problem 20 Show that the sum and product of two simple functions are simple. We know that ( . D is dense. x A or x B QED. Proof We wish to show that {x : f(x) > } is measurable } is measurable . Show that f is measurable. so {x : f(x) > } is measurable . 1 Consider n 1 An }. ) n D}. if x A B A B= 0. if x A B 1. . Take D then define An = {x : f(x) > n. D is equivalent to the condition {x : f(x) > with n ( 1. ) n D is nonempty n since. if x A and x B A B= 0. thus A B A B. if x A or x B But x A and x B x A B and.nb 2 Problem 19 Let D be a dense set of real numbers. A B A B x A B. Proof 1.MATH6303. n < . {x : f(x) > } is measurable.
Since B and C are equivalent to f restricted to D and E respectively. . Thus if A is measurable B and C are measurable. if x 1 1 A= 1 1 Thus A A A 1 0 0. we must have that f is measurable if and only if its restrictions to D and E are measurable. x D . if x A= 1. if x 1. so C would be measurable. x D . x E =B D E is measurable . if x A or x B 0. if x B A B A B A B A A B B B = B A A B A B = But x A A A or x A B B x A B . QED. x E x: f x Clearly if B and C are measurable then A is measurable as the union of measurable sets.nb 3 A A A 1.MATH6303. Show that f is measurable if and only if its restrictions to D and E are measurable. we have that A ~ D is measurable if A is measurable. QED. if x B= 0. Proof f is measurable A x: f x x: f x . C x: f x . but A ~ D = C. if x A and x B B and. 1. and similarly A ~ E = B gives B measurable. if x A . thus 1 A 0. Since D is measurable. x A and x B x A B. QED A A Problem 21 a) Let D and E be measurable sets and f a function with domain D E.
QED Problem 23 Prove Proposition 23 by establishing the following lemmas: a) Given a measurable function f on [a. < 0. and given > 0. Proof We know that A = x : f x Thus C = A and B = x : f x . QED. n such that mC 0 < = 0. We also know that C since < Furthermore we know that since f is infinite on only a set of measure 0. + 1. for some n. x : g x } = x:f x D . {x : g(x) > measurable. } is measurable iff f is 0.MATH6303. However the first term on the right is clearly measurable iff f is measurable. x:g x is measurable iff f is measurable. B is measurable = x : f x C 1. thus {x : g(x) > = x:f x . x D x :x D. but D is measurable since D is measurable. we have that m n 1 C 0 n = limn mC definition of limit. b] that takes the values ± only on a set of measure 0.nb 4 b) Let f be a function with measurable domain D. x D is measurable . with mD < / 3. Proof f is measurable x: f x . 0 0 are measurable for all . x . then . Thus by Proposition 14. D = C 0 n . Hence by . and the second is just x D. so g is measurable iff f is measurable. Thus x : g x is measurable for < 0 iff f is measurable. M such that f M except on a set of measure less than / 3. Show that f is measurable iff the function g defined by g(x) = f(x) for x D and g(x) = 0 for x D is measurable.
bm . Proof Let be a simple function with values of n*( / 2) where n ((M+1)*2/ . Given > 0 and M. If m M then we can take g such that m g M. And we know that limit Kn 0 so. Thus we have defined such that it has the neccesary property. This construction will clearly yield results of the form m M.2. x am. Thus simply choose the value for n that satisfies f(x) . [0.(x) < when f(x) < M . . However Kn forms a strictly decreasing series since for successive n. bm)). with m partition [a.1..f(x) < . (M+1)*2/ ). Also we have that 2n m({x : g(x) = (x)}) = m 1 m(Ai [am. bm] with Ai = {x : (x) = m i }.1 2 + 2 * f(x) / > n > 2 * f(x) / . when m f M.nb 5 b) Let f be a measurable function on [a. 2n ]. bm)). c) Given a simple function on [a. If m f M. each pair of new partition pieces must cover at least as well as the original. Such an n being guaranteed to exist by the fact that at least one integer in every range 1 + a > x > a . b]. Hence Kn = m({x : g(x) (x)}) = 2n m 1 m(Ai [am. Clearly g(x) is a step function and in the limit n we have that g(x) = (x). bm)) is the max over all i. bm] = [a + (m1)* . Proof Let = b2na .(x) < f(x) + + f(x) > (x) > f(x) + f(x) > n*( / 2) > f(x) 1 + f(x) / > n / 2 > f(x) / . b] into n intervals of width . b]. a simple function such that f(x) . .MATH6303. Thus some step function g conforming to the same bounds as that will have g(x) = (x) except on a set of measure less than / 3. a + m* ]. n such that Kn < / 3 > 0.(x) < except where f(x) M. for the Ai associated with m. except possibly within / 2 of the bounds where we may choose the bound itself to satisfy our criterion. Thus [am. We know that (x) = Define g x m N i 1 i * Ai . with = i such that m(Ai [am. b] such that g(x) = (x) except on a set of measure less than / 3. define [am. a step function g on [a. then we may take so that m M.
MATH6303. This is true . b].. Thus fn x fn x n 0. < xN such that for each interval (xn . a partition x0 < x1 < . QED. . a continuous function h on [a. N1]. which does not exceed its boundary points. n vn x . g(x) assumes only one value. because if we were to remove any element of A. Thus > 0 we have mK = > . Thus B A = (. Hence B = (N* . xn 1 ). ) is the smallest set such that x A and n N. K. Let n x be any continuous function joining g(xn 6 N 1 ) and g( xn 6 N 1 ). Define h(x) = g x . If m g M then we can take h such that m h M. we would add an element to A with fn x Hence the measure of any set. which has total measure 3 NN 1 < 3 . Take > 0. N* ) is the collection of all x such that fn x N* ] [N* . Problem 29 Give an example to show that we must require mE < Example Let E = (. Then fn x n N. if x xn 6 N 1 . . b] such that h(x) = g(x) except on a set of measure less than / 3.nb 6 d) Given a step function g on [a. fn x . if x xn 6 N 1 Thus h(x) is neccesarily a continuous function and h(x) = g(x) except for N intervals of length 2 6 N 1 . Thus by explicit construction we have found aa continuous h satisfying the conditions. Hence showing a counterexample. ). we may construct a continuous curve joining g(c) and g(d). some N implies that  x  < N* . Since g is a step function. for n [1. Let fn x 0 x n in Proposition 23.. Proof Clearly for any interval [c. with the neccesary property of Proposition 23 will be larger than mA = ... d].
nb 7 Problem 30 Prove Ergoroff's Theorem: If < fn> is a sequence of measurable functions that converge to a realvalued function f a. then given > 0. E ~ A that fn converges to f since  fn x m. So it . QED when fn (x) restricted to A. and all n Ni . Proof Define Ai E for i with mAi < i = 2 i ! and a smallest Ni such that fn x f x  < i = 1 . Thus we have fn converging with some Nm depending only on demonstrates uniform convergence. Thus mA Am = Am mAi < x 2 i ! = !.e.  Since A = ~ f x < m 0. x Ai . on a measurable set E of finite measure. a subset A E with mA < such that fn converges to f uniformly on E ~ A.MATH6303. i Let A = Ai . We know that Ai exists by Proposition 24.
x is rational Then R b a f x b x b a and R b a f x x 0 Proof R a f x x inf a x x. x.Chapter 4 by Bobby Rohde 92800 Problem 1 a) Show that if f x 0. and in the domain each open interal an rational number since the rationals are dense. Thus (x) f(x). when (x) = 1. (x) 0. x implies that the value over any open interval of the step function (x) is 1. QED . b a sup b a step functions (x) f(x). step functions (x) f(x). Thus (x) f(x). x. x is irrational 1. Since irrationals are dense. But each step function is composed of a finite collection of open intervals. each interval in the composition of has an irrational in its domain.Real Analysis . Thus the supremum of the integral of all such step functions must be when (x) = 0. b a must 1 be b a. Hence R b a f x x sup b a x x 0. x that over any open interval in . x. Thus the infimum of the integral of R R b a b all f x f x such x x inf step b a functions x x x x.Math 630 Homework Set #4 .
1 0 fn x = n 1 n . Riemann integrable functions such that fn increases monotonically to f. So Thus we have that the limit of the integral is the integral of the limit.MATH6304. . limn 1 0 fn 1 x 1 1 0 f so limit n x 1 0 limn we have f = 1. fn x.nb 2 b) Construct a sequence < fn> of nonnegative.1]. What does this imply about changing the order of integration and the limiting process. Construction Let fn = x n over (0.
So h( ) > ( ). and let h be the upper envelope of f.nb 3 Problem 2 a) Let f be a bounded function on [a. Thus limn n x = h(x) and clearly n is bounded since h and f are bounded.h( )  < / 2 and 0 <  x . when f(y) < limx y f x . however this works > 0 which implies that must be a boundary point between intervals of the partition. in that interval with h( ) h(x) + .MATH6304. Thus h except on a set of measure 0. in that range such that f 2 h 2 f 2 h 2 f h f h h f . However h(x) lim x h . Define < n x > to be a sequence of step function such that n is over a equipartition of [a. < . However possibly where f(x) is not upper semicontinuous hy inf sup f x so h(y) = limx 0 x y y f x . That implies that > 0 such that n. step functions (x) f(x). x. 00 x y b b b f a b a h. x+ ). BWOC Suppose limn h(x). We know from Problem 2.51(a) that h(x) f(x). Let A = {x : (x) < h(x)} Suppose A .51(a). so (x) h(x). and h( ) = limx f x .51(b) we have that h(x) is upper semicontinuous so that h(y) limx y h x = inf sup h x . However since is a step function it is composed of a finite number of distinct intervals and thus has a finite number of boundary points. By definition of lim sup we have that > 0 such that  sup f(x) .. which gurantees that for some small distance > 0. Thus by . with the value of n over each piece to be sup h(x) over that piece. Then R Proof R a f x x inf a x x. Thus we have a contradiction since h( ) must be < h(x) + for within of x. b]. From 2. h x f x if and only if f is upper semicontinuous at x. and from 2. h(x) + > h( ) (x.h(x)  n x partition intervals including x.  n x . Which means that and must be in differant intervals of the partition of . b] into 2n pieces. Let A. except f(y) < limx y f x . Let = h( ). So A has at most a finite number of items in it. so R b f x a x = inf b a x x b ah x x.( ).
.51(c) gives us that g(x) = h(x) if b ah and only f is continuous at x. b a f b a h. x we also know that h h x g x has measure 0. b a However 2.nb 4 Proposition 6 we have that so that limn Hence b a h b a n x b b ah b inf a x = limn x=R b a b a n x. So using the previous bound we have R f a b a h. f f x x. QED. By symmetry f b a g. of defintion. Furthermore n are each step functions f. interchanging the supremums and infimums we know that R where g is = b a g. where h is the upper envelope from (a). But the set of points h x g x is identical to and hence a h g the set of the discontinuities in f. g 0 if and only if. So f is Riemann integrable if and only if b ah b a b a f =R b a f. the set of points were > 0 such that m( x : h x g x ) > 0. So by Proposition 5(i). b) Use Part (a) to prove that: A bounded function f on [a. *(measure of that set) > 0. Proof A bounded function f is Riemann integrable if and only if R Since f is bounded we know R the lower envelope. QED. we know that But since h x g x .MATH6304. where is any step function b x R f x a x. since otherwise some b = b a g b a h g 0. f is Riemann integrable if and only if the set of points at which f is discontinuous has measure 0. b] is Riemann integrable if and only if the set of points at which f is discontinuous has measure 0.
then A Bf Af Bf Proof f f A B However from (i) we have that A Bf A B A . since B A B = A A B. so E A A mE and B mE. so E f g 0 Ef E g. since it exists it must equal sup E . EB (v) If A and B are disjoint measurable sets of finte measure. f g 0 a. then Ef Eg Hence Ef E f 0 f g 0. However the integral of a constant is just that constant times the measure of the set integrated over. QED E f E f (iv) If A f(x) B. It follows immediately that E f E f Ef.e..e. Proof Ef Eg Ef Eg E Thus for simple functions f g 0 a.(v) of Proposition 5 (iii) If f g a. QED. However we know that the integral must exist since the integral of f and g exist. E f Ef. f = f f B Af Bf . then Ef A mE B mE Proof Since A f(x) B. However f g a. Since 0 except on a set of measure 0..nb 5 Prove parts (iii) . sup E 0. Hence A m E E f B mE.e. But also  f  =  f  f.MATH6304.e. Ef since  f  E f f. QED. we know from (iii) that E A Ef E B. so Ef E f .
Chapter 4 by Bobby Rohde 100500 Problem 3 Let f be a nonnegative measurable function. Show that that f 0 a. x. then mA > 0.Real Analysis . which implies that f . Define h x We know nonnegative f F min f x . Thus h x f f F 1 N F f x. such that 0 < mF < . we know that the countable sum of the measures of the sets on the righthand side is greater than or equal to the measure of the left hand side which is non1 zero. Let F A. x. g g. which This implies the f sup f g h mF Thus 0 a.e. x E. Proof BWOC. QED contradicts assumption that f 0. and f F f . the latter coming from the fact that since f is with 0.e. f 0. f sup f F F 1 N . f F and g f. x. suppose f 0 a. Then f 0 implies some set E with mE > 0 such that f x 0. 1 0 x f x Since x f x n n .Math 630 Homework Set #5 .e. So N such that if A = x f x N .
2 . Now we need only show that f = lim n. 1 1 1 we know that n x f x to the spacing of the values of n = 2n < 2 log 1 = . So f = lim n . and log2 x = to the logarithim base 2 of x. we know that 2n 2log2 f x0 f x0 . x. m 2 0 2n 2n assigning a nonzero value to n x0 . n so x0 meets the first condition for assigning a Since n > [log2 f x0 ] > log2 f x0 . This implies that n 1 x 2 m 1 m 1 m m 1 m 2m f x f x 22 n . a) Show that an increasing sequence of nonnegative simple functions each of n which vanishes outside a set of finite measure such that f lim n .MATH6305. thus suppose n x = 2n for some integer m with 0 < m 22 n . Thus nonnegative simple function and vanishes except on a set of finite measure. [log2 f x0 ]}. So m 22 n n . n. We wish to show that n > N. f x0 < . n x n converges pointwise to f x. m 1 2n Define n = . but if m 1 2 m 1 m 1 2 22 n 22 n 1 . Construction m f x m with m 22 n 2n and 0. Thus f x < . for some m 2 n . log2 . then clearly 2n 2n 2n 1 2n 2n 2n 1 . n x0 Since n > x0 nonzero value to x0 . n such that m 1 2n n is a x and n as follows. so its an 2n 1 2n increasing sequence. Thus x meets the second condition for f x0 2 . with log2 0 = Given > 0 and some x0 . Finally since n > log2 ] and n meets the requirements to assign x0 a particular nonzero value.. Suppose n x = 0 then clearly Furthermore we can show that n 1 n m 1 0. we have that x0 n x0 .nb 2 Problem 4 Let f be a nonnegative measurable function. so thus we have that n 1 x 2 m 1 n x . otherwise Thus n assumes exactly 22 n 1 nonzero values over a set of measure 2n. Let [x] denote the least natural number . we may choose N = max{[ x0 ]. x n.
0 . f.nb 3 Thus we have constructed a b) Show that Proof By definition finite. 0. Thus by Proposition However f. f sup over all simple functions sup f h. an increasing sequence such that n lim n h' x . QED. The measurability of h' x deriving from the fact that x : h x 0 and the set x : h x 0 must both be measurable. Thus the sup over some subset of all such simple functions must be = h sup f h= f . over all simple functions f. Hence we may apply part a) and conclude that h' x . Thus by Proposition 10 we have that h' n which equals the sup n since it is an increasing sequence. f h n satisfying all required properties. QED . f sup . However all such h' x are nonnegative measurable functions. So f sup .MATH6305. where h is a bounded measurable function with m x : h x 0 is We may of course limit ourselves to taking the sup over all h' x such that h' x max h x . and since we are considering a sup we may assume that 8 we have that f . So f sup .
x < < 1. Hence a f = limx a x f Fa = . We need f = limx a 1 f x. by Proposition 12. otherwise a). x a n Define fn then fn is an increasing sequence with fn f as n x 0. Let N(x) be the x maximum of f y y [a. if a x1 x2 . thus as a Thus limx a f limn fn . we have that x x a x f. x a a x a f a f. Hence. Thus without loss of generality we may assume that f < everywhere since this won't change it's integral. x. A.nb 4 Problem 5 Let f be a nonnegative integrable function. Since if f = on some set. we would have f mA . Show that the function F defined by F x x f is continuous by using Theorem 10. f . Proof In order to show continuity we need that F a = limx a F x . and by Theorem 10 we have that limn x a also we know that x such that 0 < a . So limx a x fn M< 1 f limn x a fn a a f.. Thus limx a limx a F x . However it is clear x that N x1 N x2 . which is false since f . a f m[a. of measure > 0.MATH6305. x a 0 in the limit x a . x]*N(x). x Now consider limx a f f limx a a f . Now since f is finite over a closed interval we know that f is bounded over that interval.e. by Proposition 5. x < a. We know that since f is integrable f < a. by definition of integrable. Thus limx a a f limx a m a. x]. we get that limx a f limn fn . f f limx F is continuous. But . (. f a a fN for some N such that x a 1 N a N> a 1 a x . x N x = 0. we have that f fM .
we know that fn f. with fn x f lim 0 otherwise. Proof By Fatou's Lemma we know that n. QED N > x. QED b) Show that the Monotone Convergence Theorem need not hold for decreasing sequences of functions. n > N. Proof Let fn x = 0 if x x. Thus fn 1. So 1 f 0. fn n 1 fn n n > N. However consider f . Problem 7 a) Show that we may have strict inequality in Fatou's Lemma. lim fn . and suppose fn f n. Hence 0. Also since fn f n. n. lim n and fn x = 1 for x 0.nb 5 Problem 6 Let f n be a sequence of nonnegative measurable functions that converge to f. such that fn x fn .MATH6305. QED So by squeezing. N > x. f = 0. lim However lim fn f lim fn lim fn fn by definition of lim sup and lim inf. Thus fn fn 1 1. Hence lim . such that fn x n. Hence lim fn f 0. since fn defined by fn x = 1 if n 0. however lim fn . Thus 0 x n 1. Proof Consider the sequence x. Then f lim fn f lim fn . we have fn 0 n 1 1 n f = 0. and thus lim fn f. fn 1.
However f = f + f . So we have shown the first part. So f integrable ( f + f ) is integrable. E f f f A B f Af B f .Chapter 4 by Bobby Rohde 101200 Problem 10 a) Show that if f is integrable over E then so is f and Ef E f Does the integrability of f imply integrability of f? Proof f is integrable f and f are each integrable. so by Proposition 15ii. but since f and f are nonnegative we can drop the asbsolute value signs.Math 630 Homework Set #6 . Thus f and f are each integrable iff f is a measurable function. Thus f implies . so E f Ef Ef E f . and A and B must be measurable iff f is a measurable function. we have that f is integrable and Also E f Ef Ef . But this implies that A f Bf A B f integrability of f iff f is a measurable function. is integrable. f f f . Ef Ef Ef Ef E f . Thus by Proposition 15.Real Analysis . f Ef . However the set A x f x 0 is disjoint from the set B x f x 0 .
WLOG I assume that a < b. n x b . Then lim fn x = f x a. be a Riemann integrable function with improper limit point a. and f defined for all x (a. Then lim fn x = f x a. Note that if any other point in (a. Thus by and hence b b a f . show that the improper Riemann integral is equal to the Lebesgue integral whenever the former exists. In particular we know from analysis that improper points must be disconnected in order for a function to be Riemann integrable. which allows us to apply Proposition 4 to get that lim R b a fn R b a lim fn R b a f. b a fn But by construction we must have that fn x is So b a f bounded. = R b a fn . but all parts of the proof may be reversed for a > b. otherwise Theorem 16 we have that fn x lim b a fn b f n .e. Proof Let R a f x. b] is improper. Case a = Then consider the sequence of measurable integrable functions fn x f n . which allows us to apply Proposition 4 to get that lim b a fn lim R b a fn R b a lim fn R b a f. Thus by 0. and hence QED . If f is integrable.nb 2 b) The improper Riemann integral of a function may exist without the function being integrable. b]. So b a f R b a f.MATH6306. such that f . But by construction we must have that fn x is bounded. we may into two or more integrals each of which with one improper limit point so we need only assume one limit is improper. b a f. otherwise b b Theorem 16 we have that a f lim a fn . b a fn = R b a fn . Also we have that fn f . Also we have that fn lim R b a fn . such that f x. and will still hold accordingly.e. a 1 n x 0. Case a < Then consider the sequence of measurable integrable functions f x.
where i 1 bi Bi and n2 n1 n2 So i 1 ci Ci .e. we can thus deduce that n1 n2 n1 sup sup i 1 bi mBi . since this is true for any lim sups and infs. Since it follows immediately from the definition of f and f that and are simple. j and each bi corresponds to some a j 0 and each ci corresponds to some Bi n1 n2 n1 n2 n aj 0. Also we know that lim fn fn n thus By Proposition 15iii. Problem 12 Let g be an integrable function on a set E and suppose that f n is a sequence of measurable functions such that fn x g x a. so i 1 bi mBi ci mCi i 1 ai mAi . Proof Consider . also we know that Cj i. must follow the same correspondence. Show that they are equal. Since the i 1 ci mCi i 1 bi mBi i 1 sets Ci and Bi . we have that E lim fn n E fn lim E fn . QED . therefore fn is integrable. and thus we are done.MATH6306. and the right hand side evaluated according to the old rule for integrating step functions. we know that nonnegative measurable functions f. f sup . n and i 1 ai mAi . i 1 ci mCi . fn is measurable and bounded by an integrable function. we have two definitions for . on E. A similar argument for lim sup fn fn gives us that lim E fn E lim fn E lim fn . E fn lim fn lim and lim E fn E lim fn . Then E lim fn E fn E lim lim E fn lim E fn E lim fn Proof We know for free that lim We only need to show that E fn . with the sup taken over all simple functions f. QED.nb 3 Problem 11 If is a simple function. i 1 bi mBi i 1 ci mCi . By problem 4b.
. x. then we may clearly choose = 0. a) a simple function E such that f Construction Since E f . Then given > 0. > 0. we know that a step function such that = a simple function except on a set of less than / 3 and if M bounds then M bounds . Let N be a number such that x N. it suffices to show that the Ef Ef E Ef Ef E latter is less than . Ef E Ef Ef A B Ef A Ef B which again all that we need to show is that this new expression is less than . This is possible since simple functions are always bounded. Therefore E f . which part of the proof to Proposition 22. Let > 0 and be a step function such that E f 2. By Problem 4. x E A B . Thus if we I I choose over A = and over B = then we have that S I. Thus < E A f + = E A f + E f E A f A f A f + A f + A f + A A f E A f E A f / 2 + 2*N *mA = / 2 + 2*N* 4 N = .nb 4 Problem 15 Let f be integrable over E. we know that nonnegative measurable functions f we have that f sup . taken over all simple functions f.MATH6306.23c. The second implies that 0 f 2. If we further specify that A x f x 0 and B x f x 0 . and bounded by N. Choose such that = except on a set of measure 4 N with the set denoted by A. 2 2 . Thus we have shown the construction as Ef A Ef B required. Since f and f are nonnegative measurable functions we know there must exist simple functions S and I respectively such that 0 f 2 S and 0 f 2. So we may further simplify the above expression to E f . a) a step function E such that f Construction By the Question 3.
which agress with the bounds on .nb 5 a) a continuous function g such that E f g Construction By question 3.23d. Thus we may construct g such that E f g .MATH6306. except that we have to infer the boundedness of from the boundedness of the preceding . With this we may replace the occurances of in the above proof with and with g and the proof will precede identically. . we have that a continuous function g such that g = a step function except on a set of measure less than / 3.
since for arbitrarily small positive h we sine its maximal value. . D f 0 = lim h 0 f h can find a value of D f 0 h h 0 1 h which gives f 0 = lim h sin h 1 h = lim sin h 0 1 h = 1. and D f 0 . D f 0 = lim h 0 f 0 f h 1 h h = lim h 0 h sin h 1 h = lim sin h 0 1 h = 1.Chapter 5 by Bobby Rohde 101900 Problem 1 Let f be the function defined by f 0 0 and f x Find D f 0 . D f 0 x sin 1 x for x 0. D f 0 . D f 0 = lim h 0 f h h 1 h f 0 = lim h 0 h sin h 1 h = lim sin h 0 1 h = 1. D f 0 .Math 630 Homework Set #7 .Real Analysis . since for arbitrarily small positive h we can find a value of which gives sine its minimal value. since for arbitrarily small positive h we can find a value of D f 0 which gives sine its maximal value. since for arbitrarily small positive h we can find a value of D f 0 which gives sine its minimal value. D f 0 = lim h 0 f 0 f h 1 h h = lim h 0 h sin h 1 h = lim sin h 0 1 h = 1.
MATH6307. then D g x = D f lim f x f h x h x = lim h 0 gx gx h h h 0 = h 0 lim gx gx h h = lim g x h h 0 gx h = . b) If g x = f Proof D f x = D g x .nb 2 Problem 2 a) Show that D Proof D h 0 f x = D f x lim = f x h f x f x 0 = f x h f x h 0 h = f x h 0 f x lim = f x h f x h inf inf h 0 0 sup inf h h 0 lim h 0 0 f x h f x = h = inf sup f x f x h = D f x . QED x . QED.
WLOG we may consider h to be sufficiently small such that c h and c h I. Consider D f c = lim h 0 f c h f c h . By definition lim g h h 00 inf sup g but as h 0 . f c h f c h 0 f c h f c h 0. h 0 Thus we have that lim h 0 f c h f c h f c 0. f c h f has f c h c h h h 0 f so lim f c h c h 0. Similarly D f c extended reals. . and if f b is a local max then 0 D f b D f b. however the other two may not even exist. since c h f c h f c h I. since lim sups are neccesarily greater Proof Well D f c = lim h 0 D f c = lim h 0 f c h f c h than lim infs. the supremums form a nonincreasing sequence. b). b] and assummes a local maximum at c then D f c D f c f c h f c h (a. b) What if f has a local maximum at a or b? Answer If f is maximal at a or b then the two derivates which are still inside the interval must exist and obey the above inequality.MATH6307. b such that c D f c I and f c is the maximum of f over I. .nb 3 Problem 3 a) If f is continuous on [a. we know that 0. QED. Hence D f h 0 lim > 0 over I. Furthermore since h > 0. Thus with a < b and f a a local maximum we have that D f a D f a 0. Since f c is a local maximum an open interval I a. and the limit must exist over the c 0 D f c . and the limit exists. So we only really need to establish that 0 D f c . Similarly D f c D f c . since each f c h f c h < 0. lim g h must exist. and hence so long as we permit ± as acceptable limits. 0 D f c D f c .
So this means that f x f y 0 and f x f y y x . where c is the location of the local max. is a nondecreasing function. b .MATH6307. then f is nondecreasing on a. but D g x which gives a contradiction. b and one of its derivates (say D ) is everywhere nonnegative on a. f x x. Thus we know that g is nondecreasing.e. yet we know that x f y y. Suppose x y. Proof Lemma . Which is a contradiction. x. . If g were always decreasing then we know that D g x = lim g x hh g x < 0 since gx h gx 0. i. h 0 However by problem 3. such that g x g y . Let = f x f x f y . Which give that *(yx) = 2 . the above property holds. Hence f x must be nondecreasing. f x f y for x y. D g c 0. However we may choose = y x 2 > 0. Consider f x lim gx h gx h 0. h 0 f x h f x h gx h gx h Therefore D f x = lim h 0 = lim h 0 gx h x h h gx x = lim h 0 gx h gx h h h = = lim h 0 0.For a function g x such that D g x > 0. Which is a contradiction. So if f is an arbitrary function of the type given in the problem then we know that > 0. then > 0. Suppose x y such that f x f y .nb 4 Problem 4 Prove: If f is continuous on a. x gx x. b . where g x is defined as in the lemma. thus g has a local max. and y x . Since g is continuous we may apply the normal knowledge of continuous functions to conclude that either g has a local maximum value or g is always decreasing.
So by definition all the derivates are 0 if D f x 0. However we know that nondecreasing is equivalent to D f x 0. QED.MATH6307. However if we replace D f x with D f x and the related definitions in the above two steps its easy to see that the remainder of the proof will preceed verbatim. so if respectively D f x and D f x are 0 than the other two will be. D f x and D f x Now we need only look at the other direction. So in order to complete the other direction we only need that D f x 0 implies that the function is nondecreasing. This completes the proof. . and it follows immediately that f x h f x and f x f x h are each 0 for h > 0. Clearly for both D f x D f x . Hence D f x 0 implies that f x is nondecreasing.nb 5 Now all we need to show is that D f x 0 iff the other derivates are 0. However this means that the terms for which we are considering the limit of in each derivate must be always 0.
The number of discontinuities of g as above is at most countable. Proof Lemma: For a monotone BV function g on a. However we can start at x a. is finite. b . b the limit of f x exists as x c and also as x c .Real Analysis . which is a contradiction. Hence the number is finite. Thus the number of discontinuities is countable. which since g is monotonic implies that n gb . . 1 Let An x g x gx > 0}. Prove that a monotone function (and hence a function of BV) can have only a countable number of discountinuities. b . So the set of all discontinuities is n 1 An . Since it is a BV.Math 630 Homework Set #8 . g x gx . function must be in some An for n sufficiently large. n . the number of discontinuies such that 1 0. n BWOC Suppose this number is infinite.Chapter 5 by Bobby Rohde 110300 Problem 7 a) Let f be of bounded variation (BV) on a. It is clear that any discontinuity of a monotone n . and proceed towards b such that each discontinuity we choose succifiently small that infintely many disconinuties between the current location and b. Thus we experience arbitrarily many increases of at least 1 . we know that g is strictly real valued. Show that for each c a. which must be countable since it is a countable union of finite sets.
QED b) Construct a monotone function on a. and h > 0. so limit h p limx p f x limx p f x 2N . Similarly for h.nb 2 The limit of f x exists as x c and x c Consider a fixed c. Suppose > 0 such that f over c . Define i 1 fri . be an enumeration of the rationals in the interval 0. So 1 f y f x fq y 2N . Construction Let the sequence fr x Let f x r. such that x with 0 c x f x f c . h. Thus f x is monotonicly increasing. Furthermore p . y 0. Thus f satisfies all the conditions desired. b which is discontinuous at each rational point. Thus we may assume that > 0 a discontuinity in f over the region in question. x r . However since p implies that f p p h fp p p. Thus since the lefthand side is a monotonically increasing Clearly i 1 fri x 2n i 1 2n bounded series it must converge to some number less than infinity. contains no discontinuties. using the above property.MATH6308. So we may take sufficiently small such that all points of differance > 2 are more than away. for some N since all terms for which fri x 0 yields fri x fri y . c . and fq x 0 fq y . However we know that f is BV and thus the sum of two monotone functions g. Thus by the first lemma there exist only finitely many points with g x g c n1 . f p h f p h f p p h 21N . 1 with x y. Hence limit x c exists. Since they contribute with opposite sign. we have that some q x. then f is continuous in that region and limit of f x as x c exists. QED. > 0. Furthermore it should be clear that x. Thus h > 0. . 1 1 1. y . We need to show that given > 0. The proof however clearly proceeds likewise for x c . were is the minimum of the two 's needed. Hence f x is discontinuous at p since the limits can not be equal. r x 1 2n ri i 1. Claim that f is such a function as required. p p f p h f p h 0. since rationals are dense. x 0. Which implies that f p h f p h 0. f x f c . 1 . 2N 2N .
If we break the new partition as above we have two partitions about a. xi 1 then consider the partition a x0 x1 . respectively with t taken over these partitions summed = c b b t1 . or c xi . b . g f xi k i 1 1 f g xi f g xi g xi 1 g xi 1 = by the triangle Thus sup t f b t g . and Ta c f c b Ta f . b b) Show that Ta f g b Ta f b b Ta g . for those partitions. Then either c xi . QED . xi 1 . However Ta = sup t over all possible partitions. b respectively can be used to form a partition over a. b into two partitions..MATH6308. c and c. for some i. where k g xi f xi 1 g xi 1 i 1 f xi inequality.. i 1 f xi However If c xi . xi c xi 1 . Showing that Ta b Ta . then Ta c Ta b c Tc and that hence Ta b Ta . In this If c partition t2 is to the original t t1 . c and c.. xi . by triangle inequality. k t f xi 1 . since f xi 1 f xi f xi 1 f c f c f xi f xi 1 f c f c f xi . Hence Ta Tc Ta . however this last term is just t f g sup t f k i t f g 1 f xi tg. where sup is taken over all possible partitions of a. then clearly we may split the partition over a. p2 over a. for some i. such that ta is equal to ta + tc . hence Ta f b Ta f b Ta g . b . b Thus Ta c Ta b c Tc . b . Proof b Ta f g b Ta f b Ta g b Ta f g sup t f g . so this includes the b c partition formed by the union of p1 and p2 and hence Ta Ta Tcb . b By definition Ta sup t. b b by joining the two at c. c and b c b the other over c. one over a..nb 3 Problem 8 a) Show that if a Proof b Ta c Ta b Tc c b b. xk b. t2 b Ta c Ta b Tc Every partition p1 .
sup c t f c k i 1 c f xi c f xi 1 = k i 1 c f xi f xi 1 Thus sup t c f b sup t f . 1 ? Answer 0 and f x x2 sin 1 x2 . thus a2 n 2 . for some integer n. QED Problem 10 a) Let f be defined by f 0 on 1. for x 0. where t c f f xi f xi 1 = c t f . Thus an adjacent maximum at 1 1 1 1 n So a2 sin a2 b2 sin b2 = n 1 1 1= b such that b2 2. n 2 2 n 1 2 2 n 1 2 2 n = 4 n28 1 . Therefore f is not a BV.MATH6308. will be = a2 sin a2 b2 sin b2 . hence Ta c f c b Ta f . b . . Over any interval a. since the last series is a constant times the n 1 4 n2 1 n 12 n harmonic series. the diffence from x2 sin x2 .nb 4 b Ta c f c k i 1 b Ta f b Ta c f = c sup t c f . 1 Suppose a is a minimum. So 8 n 1 0 f T 1 . However as the number of partitions gets arbitrarily large we may add arbitrarily many terms of this form with n increasing. Is f of BV 1 1 1 No.
So we only need that n 1 2 n2 1 4 2 n4 2 n2 n 2 1 4 4 0 0 Ttail. where Ttail is the part not < . On inspection it is clear that t over a monotonic region will be independant of the choice of partition and have the value equal to the differance of the end points. x . Since g x is symmetric in x we know that it is suffiecent to show that g x is BV on 1. Is g of BV Yes. QED. It is obvious that b . which is a convergent sequence. It is clear that it will proceed monotonically in this region. b From problem 8a) and an obvious induction it is clear that Ta may be broken into finitely many pieces. purely by inspection and our knowledge of basic calculus. Thus our sum must converge. n 1 1 . Now we want to consider the total T 01 covered in m steps. However this is 3 bounded above by n2 for n large. and the proceeding m2 pieces follow for decreasing values of n. for x 0. the tail contribution must go to 0. However this may be rewritten as . So n n 1 2 2 sin n 2 n 1 2 2 2 2 2 2 n 2 2 2 < n 1 2 n2 1 4 n 12 n 2 1 2 2 2 sin n 1 2 n2 n 2 1 4 4 2 = n 1 2 2 2 n 1 1 2 2 2 = < .MATH6308. Now we need to evaluate T over one these pieces that have been set up to run from a local max to a local min or vice versa. Taking limit m the n 1 2 n2 n 2 1 4 4 T 1+ 2 n m 2 1 . and the last piece represents the tail going to 0.nb 5 b) Let g be defined by g 0 on 1. Consider a single transition for a minimum to an adjacent maximum. So the points in question have the form n 1 . So T 11 x2 sin 1 is BV. Hence T 01 . 0 . 1 ? Answer 0 and g x x2 sin 1 x . Proceed by breaking Ta into m pieces such that the first piece is T 1 . (Either p or n is 0). We know that minima and maxima occur at sin T 1< 2 2 1 x 1 x= n 1 2 . note that the inequalities depend on the fact that n < 0.
MATH6308. 2 . b . Further n i 1 1. Let M > 0 be a bound on f . 2 g xi g xi f xi g xi < 2 n i 1 f xi 2= . Proof Let f and g be AC. i 1 xi Consider n i 1 f xi Thus f 2 n i 1 f xi f xi 2 f xi f xi f xi 2 n i 1 = f xi n i 1 f xi f xi f xi 2M < 2M f xi f xi 2M = . Since f is AC. b is a closed interval. . Concluding the Lemma. Since f xi f xi = f xi f xi that f is AC. 2 of such that intervals n i 1 f xi xi. . i 1 xi n However i 1 f xi inequlaity. and hence that since a. is AC. >0 n such that f xi 2 M . finite collection of intervals xi . Then we know that f is BV. g xi g xi . xi . f xi such 2 and that g xi f xi min f xi f xi 1. by triangle Hence we may choose in order to satisfy the condition and show that f = f xi f xi g is AC. Then n g xi 2 finite collection i 1 g xi n xi min 1 . Let > 0. such that i 1 f xi n xi . Let f be AC with respect to a. that f is bounded. xi . Let > 0. we know that f is AC implies b) Show that the product of two AC functions is AC Proof Lemma: f is AC implies that f 2 is AC.nb 6 Problem 14 a) Show that the sum and differance of two AC functions are also AC. So f g must also be AC.
MATH6308. But for c f . f g f g f g . so we 2 only need that constant multiples of AC functions are AC. Proof Since f is never 0 over a closed interval. differances and squares of AC functions are AC. we know that m > 0 such that f n > 0. c) If f is AC on a. QED. Since f is AC. . b . (Note c = 0 is obviously true. Let = n i 1 f xi f xi f xi f xi n i 1 f xi f xi m2 < m2 m2 = . is AC. > 0. and the c will carry through to make c f . QED. but that both sums. Consider Thus g 1 f n i 1 1 f xi 1 f xi 1 f is m everywhere. so it is also AC.nb 7 If f .) 2 2 2 Hence f g can be gotten as a sequence of operations that perserve AC. g are AC over a. such that i 1 f xi f xi m2 . b then f g is AC. b and if f is never zero there. then the function g also AC on a. where f is AC it follows immediately that for any > 0 we may choose f constrained by c . AC.
we denote l p the space of all sequences v v 1 such . However expnading this expression and summing we see that it is in turn = 1 p p So v 1 v + = v v p+ v p . > 0 such that p p v 1 v 1 v v p p 1 p v v 1 v v p 1 p v p v 1 1 p = p v 1 v v v p v 1 p = 1 p p p . Prove the Minkowski ineqality for sequences p v p v p. If either v or v has norm 0. . QED.Real Analysis . 1 and then define real numbers . with 1.Math 630 Homework Set #10 . since p 1 and the quantities inside the distribution are each less than 1. . v p and v sup p Proof Case 1 v p p v 1 v p 1 p v v p = v 1 v v p 1 p . the proof is trivial so we may assume that each has norm greater than v v 1.Chapter 6 by Bobby Rohde 111700 Problem 7 a) For 1 p that v 1 v v v p . Here we have 1 v p p p v 1 . 1.
By Lemma 3 p t Summing v v and v v v are positive since the norm is dependant p p 1 p =p t v t v v v v 1 p v both t v pp sides. Thus 0 N such that n N. fn 0. then Proof Case p v 1 v 1. and hence f is in L . Proof fn f in L imples f n f except on a set E of measure 0. v v b) Show that if v 1 v v l p and p v v l q with q 1. except on E. If fn uniformly converges to f except on a set E of measure 0 then and n N. > 0. fn x f x . . Case 1 p q Define v p . QED. f x < fn x bounded a. Thus f is .MATH63010. v p p v 1 p t p t v p v p p p v p = v t p v v v p Differentiating with respect to t we have that v p v q . QED.nb 2 Case p v v v sup v v p sup v v sup 1 p 1 q v + sup v = . We may chosoe to assume that only on their absolute value. Prove that f n converges to f in L if and only if there is a set E of measure 0 such that fn converges uniformly to f on E. v 1 p p 1 p = Problem 10 Let f n be a sequence of functions in L . q v v 1 v sup v = v 1 v v = v 1 v . Thus fn is uniformly converges to f on E. f n f f .e. N such that x E ess sup fn x .
Proof In order to show this we need that every Cauchy sequence of functions some f in L . Since fn fm is bounded a. However if we consider the collection of all such M m > n. by . since if fn converges then it must converge to f a. . Thus fn x f x is bounded by some ess sup . if the limit exists . Since this is L fn x .e.MATH63010.nb 3 Problem 11 Prove that L is complete.e. except on a set M of measure 0. Then where the limit exists. by .e. Thus we only need to show that the limit must exist upto a set of measure 0. f is bounded by + . . Since fn is in L it union is also a set of measure 0. Define the pointwise limit f x It suffices to show that ess sup f limn fn in L converges to . otherwise . N such that m. n N. Given > 0. we have a countable collection of sets of measure zero so there if limn fn x exists. f n fm norm we know that fn fm is bounded a. by definition of Cauchy.
Clearly by many An of length < . 0. Show that to a function f in L f n converges to f in L p if and only if fn f . However. Thus showing the counter example to Proposition 8. N Consider intervals of the form An = n1 . 1 .e. 1 Hence f 2 . 1 p . Proof If fn fn fn p 1 p fn f p p f 1 p fn p f p 1 p fn 0. since p f 0 fn p 1. Show that f . x 0 Clearly f x is bounded by 1 and thus in L . Define f x sin 1 . f 2 over An .nb 4 Problem 16 Let f n be a sequence of functions in L p. 0 1 2 . g functions . fn f p ?????????????????????????????????? Demonstrate Proposition 8 fails at p = . for p 1 fn f 0 fn f p 1 p fn converges to f in L p f p. n 1 . 0. f L . Let denote the size of the smallest interval in the partition. . 1 . Clearly An 0. so regardless of the value of over m.. m 1 and length An 1 construction f An 0. So choose n. for n . m 1 .. and infinitely 1 . x 0 x . However f p f .MATH63010. Consider a step function on partition 0 1 . Counterexample on step functions. such that step functions 0 and g and continuous 0. p . which converge a. m such that An m .
Define g x 0. Consider a continuous function 1 >0 > 0 such that x 1 . QED. By continuity at x . Thus g = 2 1 2 and the interval 1 2 . 2 1 2 0. 1 2 1 2 Clearly g is in L . and the interval .MATH63010. Thus showing the counter example for Proposition . 1 2 where x 1 2 .nb 5 Counterexample on continuous functions. x 1 2 1 2 . then choose . approximating g. x 2 2 Suppose that 1 0. x 1. yet g = 1 . Suppose that So g 8 at p = 1 2 1 2 . > 0. then choose = 2 1 over this interval so g 2. 1 2 where x 2 1 2 .
. Thus .Chapter 6 & 11 by Bobby Rohde 112500 Problem 6. QED. that implies that m x:g x Hence Case f g = g sgn g = g 1 1= g 1 f . Show that 0 and measurable function f such that f f g g 1 a bounded f Proof Case Let g 1 0 sgn g x . 0. g = g . where sgn 0 0 ess sup f(x) = 1. QED 0 0. 0 . Then f f = 0.21 a) Let g be an integrable function on 0. 1 .Math 630 Homework Set #11 . f g = g g 0 finite .Real Analysis . 0 Since g 1 f x 0. 1 Let f x f g= g x.
Proof . E Let g = Problem 6. 0. QED. Let g by definition of E. Hence E g f 1 . Problem 6. we have this = mE g 1 .22 Find a representation for the bounded linear functionals on l p. 1 Construction p .nb 2 b) Let g be a bounded measurable function. Thus mE > 0. Thus f g g f t M 0. Show that integrable function f such that f g g f 1 >0 an Proof By definition ess sup g(x) = inf M : m t : f E x:g x g . Given > f x Then f g = E.24 Show that the element g in L p given by Theorem 13 is unique.MATH63011.
MATH63011.nb
3
Problem 11.7
Prove Proposition 4: If X , , is a measure space, then we can find a complete measure space X , 0, 0 such that i. ii. E iii. E Proof
First we must show that the 0 defined by (iii) is in fact a algebra. Thus we need that E1 , E2 E1 E2 0, 0 and E1 0 and the union of any countable collection of sets in is in 0 .
0
0. 0
E E
A
0
E
B where B
and A
C, C
, C
0.
Clearly E1 E1 A1 B1 where B1 and A1 C1 , C1 , C1 0. And similarly for 0 E2 . Thus E1 E2 = A1 A2 B1 B2 . And we know that B1 B2 , since is a C1 C2 . C1 C2 , since is a algebra and algebra. Furthermore A1 A2 0 C1 C2 C1 C2 0. So C1 C2 0. E1 = hence A1 B1 B1 = A1 B1 = C1 C1 . Also C1 A1 A1 B1 B1 B1 C1 . However B1 C1 , and C1 . Hence E1 has the appropriate form to be in 0 .
Thus we need only show that the union of a countable collection of sets in 0 is in 0 . Consider An Bn = Bn An , we know that Bn , so we only need 0 n 1 En , En that An is the subset of some set of measure 0. However An Cn , with Cn 0, and the union of countable collection of things of meaurse 0 is still measure 0. Hence it suffices to consider the union of the Cn and we are done. Thus 0 is a algebra. A B as above so that A is a subset of a set of measure 0. We wish to Consider E 0 with E show that B is not dependant on the choice of B . Given two such representations A1 B1 A2 B2 . Let A1 C1 , with C1 0 and A2 C2 , with C2 0. Thus, B1 A1 A2 B2 B1 C2 B2 = B2 , however by symmetry of argument, we have B2 C1 B1 = B1 , and hence B1 B2 . Thus we may define 0 E B, without concern that the value will change depending on our choice of B. So that 0 A 0 if A C, C 0.
MATH63011.nb
4
It then follows immediately that 0 is a measure from the fact that is a measure and that we have only added sets which have measure 0 and the sets generated by their unions with sets of , which will neccesarily preserve the additivity of the measure. Thus we have created a complete measure with the neccesary properties.
Problem 11.10
Prove Proposition 7: Let f be a nonnegative measurable function . Then a sequence of simple functions with n 1 lim n such that f n at each point of X. If f is defined on a finite measurable space, then we may choose the fucntions n so that each vanishes outside a set of finite measure. Construction
Define En,k x:k 2 n pair of integers n, k . I wish to show that
n
n
f x
k
1
2
n
and
n
2
n
22 n k 0k
En,k ,
with respect to each
thus defined has the neccesary properties.
2 n
2 . Thus m k 1 En,k , and so Since f is finite, for a given n, k , n 0, k 0, mEn,k 0, except on that set which is finite. So each function vanishes off a set of finite measure.
n
=
Furthermore it should be clear that for a given n, the sets En,k are mutually exclusive and that if for a given n and k, f x is bracketed by k0 2 n , k0 1 2 n then for the next n, we will have f x bracketed by either 2 k0 2 n 1 , 2 k0 1 2 n 1 or 2 k0 1 2 n 1 , 2 k0 2 2 n 1 . And the lower bound of the interval is the same as the value returned by the sum, it is then clear that n 1 n. Furthermore, since the width of the interval between k 2 n and k 1 2 n goes to 0 as n 0, while the range goes to both 0 and as k 0 and k 22 n . This allows that for any value of f x , it will eventually be in some interval and the value can never exceed f x . Thus since the interval successively approximates f x with an width equal to that of the interval hence the limit must go to f x .
MATH63011.nb
5
Problem 11.17
Prove Proposition 15: If f and g are integrable functions and E is a measurable set, then i. E c1 f c2 g c1 E f c2 E g ii. If h f and h is measurable then h is integrable. iii. If f g a.e., then f g. Proof
E
c1
f
c2 g
c1
E
f
c2
Eg
Simply break f and g into + and  parts and apply Proposition 13. If h
f and h is measurable then h is integrable.
If we can first show part (iii), then by applying that result we can conclude that f h f , and thus h in integrable. So let us proceed to part (iii). If f
g a.e., then f g.
If we consider
n
f
inf
f
, then clearly
a sequence of monotone increasing simple functions
which converge to g which will be less than f a.e. By montone convergence theorem lim
n
g
inf
f
f . Hence
f
g. QED.
then measurable. since En is i 1 Ei. i 1 Ei . so N A E A A Ei . Hence we know for any finite sequence this property will hold.Math 630 Homework Set #12 .Real Analysis . Suppose Ei has a finite number of nonempty sets. However A A i 1 Ei i 1 i 1 Ei i 1 Ei .Chapter 12 by Bobby Rohde 112500 Problem 12. yet we know form i 1 the properties of outer measure that A E A Ei . The left hand side is now independant of N so i 1 Ei i 1 we make take the right hand side to infinity. conforms to the property that i 1 n Consider E A E = A E En A E En . However E En . Proof If Ei has only one nonempty set element then this is trivial. A A Ei . Hence i 1 A E A Ei . So N N N N. and exploitning the fact that each Ei is disjoint. QED.2 Assume that Ei is a sequence of disjoint measurable sets and E = Then A we have that A E A Ei Ei . we arrive at n A E = A En A E = i 1 A Ei . Thus the induction is proved. . So A E A Ei . then by the assumption of the induction hypothesis say that n 1 n 1 E A E A Ei .
In conclusion. Thus all elements of may be expressed in terms of the union of a countable dijoint collection of sets.nb 2 Problem 12. which preserve unions. . Then has a unique extension to a measure on the algebra generated by if the following conditions are met: i) If a set C C Ci ii) If a set C in C Ci is the union of a finite disjoint collection Ci of sets in . By Problem 5. such that Ci A. A Ci .Let be a semialgebra of sets and a nonnegative set function defined on with 0 if . Proof A Ci by condition (i). However it is always possible to generate a disjoint sequence of sets from an arbitrary sequence of sets. A D j . Ci D j. property (i) gives us that the notion of measure on is well defined and unique on finite sets. for any disjoint sequence Ci .4 Prove Proposition 9 . but also by condition (i) we have that transitivity of equality. then a) Condition (i) implies that if A is the union of each of two finite disjoint collections Ci and Di of sets in . as we can on a semialgebra. we have that all sets in take the form of countable unions of sets in . Which is exactly what is required to show that the measure is countably additive. then is the union of a countable disjoint collection Ci of sets in . so be the b) Condition (ii) implies that Proof is countably additive on . and property (ii) gives us that countable additivity is preserved and that the value of A is still uniquely defined over countably additive unions. This implies that condition (ii) is equivalent to A . then Ci D j.MATH63012. provide that we may take finite intersections and complements.
respectively.nb 3 Problem 12. = . However Hence for each element in . and may be replaced in theorems n i 1 Ci . i 1 Ai i 1 Bi = i 1 Ai j 1 Bj = i 1 j 1 Ai since is closed under finite intersections.MATH63012. ni Consider A = n 1 Ci . Proof .5 Let . By induction is closed under finite intersections. but k 1 Di. yet no other elements have been added besides these. Thus since is closed under finite intersections from A i 1 k 1 Di. QED. Ci .k . such that A i 1 Ai . with Di. so that by and . A = = . Hence is an algebra of sets. where Ci . we must have that A . That it is the smallest algebra of sets containing is clear from the n fact that any algebra of sets must contain the elements of the form i 1 Ci with Ci . Proof be a semialgebra of sets and the smallest algebra of sets containing n i 1 Ci a) Show that is comprised of sets of the form A = with Ci . contains all elements of the form A i 1 Ci .k .k above. thus . However by definition of semialgebra. and B i 1 Bi . We need to show that this is sufficient to be closed under finite intersections and complements. Thus . Thus i k 1 Di. B . n m Let A.k ni ni n . Thus i 1 Ci . but each Ai B j . Thus A B is a finite union and hence an element of of the appropriate form. Clearly it is neccesary that be closed under finite unions and thus it will contain finite unions of elements in . Then n m n m n m A B B j . b) Show that . where each set Ai and Bi is in .
. Given B . then a sequence Bi then the result is trivial. C the elements of a generating sequence for C. Then we want to show that B C . and thus taking the countable union will also be in . for example.nb 4 Problem 12. By induction we may intersect any finite number of elements in with elements of . Thus the intersection of any element of with one of is in . j i Bj k 0 Bk . Given any element x B. x Bk . . however we know that B Ci from above. a) Show that Proof Well by definition is the set of all sets generated by countable unions of sets in so all that remains to be shown is that it is closed under finite intersections. with A0 hence x Ai . This is equivalent to B Ci . Given B with B . Given y B. b) Show that each set in sets in Construction is the intersection of a decreasing sequence of Clearly if the set in question is a member of term in the sequence to be that set. i N. We know that there must Bi . since is closed under finite intersections. i 1 Define Ai Bi . where Ci are Given two elements B. Thus we have a countable intersection of decreasing sets in A that yields B A . k. with each set in such that Bi B. Define Ei to be the sequence of sets such that Ei Bi C . is closed under finite intersections. B. then we know that B C B Ci . i. y BN for some N and hence y Ai . since it appears in the exist a sequence of sets closure of with respect to unions.6 Let be a collection of sets which is closed under finite unions and finite intersections. If N is the least such N that works then y Ai . is closed under countable unions and finite intersections.MATH63012. However the Ei Bi C B C.B . an algebra of sets. i N. C . and thus Ai 's form a decreasing sequence and Ai B. since we just take every i 0 of sets in such that Bi B. C .
7 Let be a finite measure on an algebra . and the induced outer measure.MATH63012. Proof . such that E A .nb 5 Problem 12. Show that a set E is measurable if and only if for each > 0 there is a set A . A E.
i 1 f ri .Chapter 12 by Bobby Rohde 121100 Problem 12. f n . I am going to say when we can apply the Fubini and Tonelli theorems and then simpify the conclusions of the theorems as applicable to this specific case. Then we can equate f x. n n x. we know that the set of positive integers is may apply the Tonelli theorem provided that f is nonnegative everywhere. y . (This measure is called the counting measure. However the right hand n side is equivalent to limn ri = i 1 f ri . Answer While I am unclear on exactly what suffices as response to this question. and let be the measure defined by setting E equal to the number of points in E if E is finite and if E is an infinite set. so we In the case of Fubini theorem we need that X Y f . since the counting measure over a i 1 f ri set of one element is 1. First considering the Tonelli theorem. Then f n . finite. Suppose we consider the positive and negative parts of f. y . Let ri i 1 be such an enumeration. y be a realvalued function from X Y. y ri x. X .19 Let X Y be the set of positive integers.) State the Fubini and Tonelli Theorems explicitly for this case. Let f x. and by limn application of the monotone convergence theorem.Real Analysis . y = i 1 f ri If we further define ri x.Math 630 Homework Set #13 . Thus saying that X Y f is equivalent to asking that and i 1 f ri . then the sequence of functions are each simple and clearly i 1 f ri converge to f. However since X Y is a set of cardinality 0 we know that an enumeration of the points in X Y.
i. y X Y f x. k of integers exactly once. and the integrals are thus the same by Monotone convergence theorem. y i 1 f ri i 1 f ri . this is identical to the above result. y . and thus A Y . i . we will only consider nonnegative functions f. y y = Y fx y = Y i 1 f x. We need only show that A is well defined for all . Which is equal to f x. i i y. (Note that technically we msut first consider this as the limit of finite sums. i is measurable follows immediately form the fact that x (X). but the functions in the finite sums are increasing nonnegative functions whose limit is the sum.5.MATH63013. and thus = i 1 j i f i. (ii) Y f x. However A Consider y : fx y is some subset of Y. so A is measurable. to find that X Y f x. = A . i i = i 1 f x. In the nonnegative case we may exploit the fact that the sum is the limit of a sequence of increasing simple function and apply the monotone convergence theorem to say that Y fx y = i 1 f x. j . X Y i 1 f ri ri 1. we will then integrate with respect to . X . (iii) X Y f x. y = X Y f x. y y is a measurable function of X . y = i 1 j i f i. However we also know that X Y f x.) However i 1 since is the counting measure. For simplicity I will not state the results generate by interchange of x and y. and hence fx y is a measurable function by Proposition 11. Let us explicitly compute Y f x. Well fx y i 1 f x. where x is fixed. y is a measurable function on i 1 f ri ri x. and = That the function i 1 f x. However since the ri 's range over ever pair j. (i) for almost all x the function fx defined by fx y Y. i . y X Y f x. that Y fx y i 1 f x. Hence X Y f x. The argument is the same as the 2nd paragraph in (i). since i 1. i . y Working from the conclusion of (ii). j i . y = ri = since ri x. .nb 2 Let us now consider the implication of the theorems which are of course exactly the same. since it is in the collection of measurable sets. Since integration is a linear operation. i . j . y = i 1 j i f i. Using the i y same caveat as above we may consider this as positive and negative pieces of f.
where the j. However H j Gk is a rectangle in X Y. X and Gi they inf g each g Y g admit a representation Gi Y . y h x g y . X Y. and define f x. g and h are nonnegative functions. y h x g y m1 n i 1 ci Ei x. y i 1 di Hi x Hi x G j y = Hi G j x. with each Ei H j i Gk i and ci d j i fk i . y . Then f is integrable on X Y and X Y f Xh Yg Proof Without loss of generality we may suppose that f . If not we may rewrite them as the differance of their negative and positive parts and by the linearity of integration we only need to show the proof for nonnegative functions. inf m1 i 1 di h h X Hi h are integrable m1 i 1 di g functions. but by definition of . We know i that each G j and Hk is in X and Y respectively since these are the basis sets. m This that n 1 ci Ei X Y = i 11 m2 d j i fk i H j i Gk i X Y . k depend on i and each pair of j. that every m2 i 1 fi Gi y = m1 i 1 is m2 j 1 di a possible function for f j Hi x G j y . Consider inf f X Y X Y f . If the integral exists then by definition of integration n i 1 ci X Y Ei f is = a Ei X Y . y n i 1 ci x. g h Since f x. so i 1 dj i H j i Gk i = H j i Gk i fk i H j i Gk i = m the j i and k i this is just equal to i 11 di m1 m2 i 1 dj i m1 m2 fk i i 1 dj i m2 Hi i 1 fi Gi Hj i = X h Gk i .MATH63013. so Yg where x. y nonnegative simple function and each Ei However Xh h h g. However This implies that f admits a representation as a sum of n m1 m2 elements. k are used exactly once over the m1 m2 terms.22 Let h and g be integrable functions on X and Y. Y g . so this devolves to m1 m2 fk i H j i Gk i . with Hi m2 i 1 fi m2 i 1 fi x x and y y.nb 3 Problem 12.
However there is nothing Xh Yg specific in this argument to the fact we choose inf.MATH63013. hence by the sup over simple functions less than the given function approach we can arrive at . but I don't know what it is so I decided to do it based on first principles.nb 4 Thus inf f X Y X Y inf h h X h inf g g Y g . except the last inequality. assuming the integral over f exists. . Author's note: I am all but sure that there is a better way to do this. inequality Xh is Yg always sup f true X Y of the sups inf f X Y and infs used to define . which implies that f exists and its value is Xh Yg . Evaluating the infs we arrive at f . QED. if the integral over f exists. X Y that middle . X Yf Xh Yg However Xh Yg what sup f X Y we have inf f X Y in fact Xh shown Yg is the integrals.
24 The following example shows that we cannot remove the hypothesis that f be nonnegative from the Tonelli Theorem or that f be integrable from the Fubini Theorem. Also at y 1. 2 3 = 2 2 1 X f x. Hence 0. y 2 2 y 2 2 y 1 = 2 y 1 1 2 y 1 . In other words I wish to show that X Y f x. .MATH63013.. otherwise 2 2 x A Contradiction First it is neccesary to notice that f x. Now in order to compute the integral over Y. y x x 2 2 Ax 2 2 Bx . each evaluation is in fact one. Hence the counting measure evaluated on these sets is just 1. So computing Y f x.2. y at a particular x. which is empty because x 0 is not in the domain of positive integers. y : x y 1 . if x y 2 2 x. note the substitution in the exponent in terms of the fixed coordinate. Again each of these has exactly one item in its set. y . where In order to disprove the Fubini and Tonelli thereoms it suffices to show that property (iii) in each fails to hold. y = N 3 3 y 1 i 1 2 = 2 1 i Since we are using the 1 1 i 2 2 1. if x y 1 0. y 2 2 x. y for a fixed y. x. y A x.. y that we first evaluate X f x. Furthermore we know that for a fixed x. Y X f x.nb 5 Problem 12. we have that Let us first consider X Y f x. Simplifying the geometric series we have .. y Now let us consider Y X f x. with the exception of B y at y 1. Thus Y X f x. Thus y y 1 gives the integral as 2 2 Ay 2 2 B y . y . y 2 2 x B x. Thus Y f x. So for y 2.N 2 counting measure. X f x. it is not hard to see that by restricing the integral to the first N integers we are in fact integrating over a simple function. Let f x. y 2. and similarly Bx has only one point. y = 2 2 x 2 2 x 0. Ax has only one point determined by x y. y . Let X Y be the positive integers and be the counting measure. y : x y and B x. X Y f x.
we have thus shown that fails to satisfy the (iii) property of the Fubini and Tunelli Theorems. Since 0 1. So we have that Y X x. QED. for a particular value of y. for a particular choice of x. So X Y x. QED. . but x and so since is the counting measure.nb 6 3 2 1 1 2 Fubini nor the Tonelli theoreoms will hold for this particular fucntion and choice of measure. in fact). y 1 2 0 X Y f x. First we start with X x. This means that Y x . y = = X 1. y. y . Then the diagonal x. we have that Y x. but its characteristic function fails to satisfy any of the equalities in condition (iii) of the Fubini and Tonelli Theorems. Let be the Lebesgue measure and the the counting measure. First we wish to compute Y x.MATH63013. Hence neither the Problem 12. 1 . Which evaluates the integral to y . y . which again is a set containing only the point where x However since is the Lesbegue measure this evaluates to 0. y = Y0 0. X1 Now consider Y X x. 1 2 = 1 2 . Contradiction Again we will seek an exception of the Fubini and Tonelli Theorems by computing the differant halves of part (iii) in each result and showing that they are not equal. Consider X Y x.25 The following example shows that we cannot remove the hypothesis that f be integrable from the Fubini Theorem or that and are finite from the Tonelli Theorem: Let X Y be the interval 0. y = y is only nonzero for the one value where x y. y X Y : x y is measurable (is an . x. with the class of Borel sets. y . y . y . y = 1. Thus Y X f x.
and hence Bx 0. but m2 .31 Let f be a nonnegative measurable function on twodimensional Lebesgue measure on 2. f x = f x . Then is a decreasing function and Proof m2 x. .nb 7 Problem 12. f x x is the same thing as the area Let A x. Finally we know that A of a characteristic function.MATH63013. Thus A = f x . y :0 y f x f x x Essentially in the first part we are attempting to show that under the curve (with and without its boundary). y :0 Let 0 m x: f x f x x. and thus that Ax f x . and let m2 be the f x f x x :0 y f x m2 x. f x . t . Hence we have that m2 A = f x x. Thus the conclusion is equally valid that m2 B f x x. For any A A particular x we have that Ax is the interval 0. since is finite and is everywhere nonnegative. we may A apply Tonelli Theorem to conclude that Ax . Then m2 x. y : 0 y f x . y . y :0 y f x m2 x. f x . y : 0 y f x and B x. but the integral with respect to Lesbegue measure is exactly what is intended by the = A by the definition of an integral notation f x x. Consider . Let denote the Lesbegue measure. If we then note that B behaves exactly the same as A in the above proof since Bx = 0. y t t t .
we have that x : f x t1 " x : f x t2 . t2 t1 . From here we need only observe that is the same 0 0 as A in the first section of this problem. upto the labelling of variables. Since characteristic functions are bounded and both measure are taken with respect to the finite real numbers. So m x : f x t1 m x : f x t2 . t t x. we may apply Tonelli thoerem to conclude that x. QED. t x t. and the observation that if x x : f x t1 then x x : f x t2 . Thus for t1 t2 .MATH63013. conclude that 0 0 . On the question that is a decreasing function. in the proof that follows we will attempt to demonstrate that it is equivalent to a series of horizontal bars. this follows immediately from that fact that mA mB. t x t = x. t :f x t 0 Consider 0 t t = 0 m x : f x t t = 0 m t t = 0 x. whenever A ! B. Let x. Thus from above we may x. t t x = f x x. t t f x x. by exploiting a series of vertical bars. The last equality coming from the fact that the measure of a crosssection is the same as the integral of its characteristic function.nb 8 0 t t f x x While in the above proof we made a construction that showed the area was the same as that under the curve f x .
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.