100%(5)100% нашли этот документ полезным (5 голосов)

9K просмотров65 страницReal Analysis Rudin Solution Manual(1-8)(10-13)

© Attribution Non-Commercial (BY-NC)

PDF, TXT или читайте онлайн в Scribd

Attribution Non-Commercial (BY-NC)

100%(5)100% нашли этот документ полезным (5 голосов)

9K просмотров65 страницReal Analysis Rudin Solution Manual(1-8)(10-13)

Attribution Non-Commercial (BY-NC)

Вы находитесь на странице: 1из 65

Homework Set #1

by Bobby Rohde

9-7-00

Problem 1

If A and B are two sets in with A B, then mA mB.

Proof

Since is a -algebra, we know that C = B A , with C A= and C A = B. Hence mB

= m(C A) = mC + mA mA. mB mA, QED.

Problem 2

Let <En> be any sequence of sets in . Then m( En) mEn.

Proof

By Proposition 1.2, a sequence <An > of sets in with En = An and An Am = , n m.

Thus m( En ) = m( An ) = mAn . If we construct An from En according to the algorithim used

in the proof of Proposition 1.2 then we have An En n, and from Problem 1 therefore mAn

mEn . Hence we have mAn mEn m( En ) mEn . QED.

MATH630-1.nb 2

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 <Ai > of sets in , n( Ai ) = nAi .

If Ai is infinite for any i then n( Ai ) = nAi = , so we may as well assume that Ai is finite i.

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 1-1 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

Clearly it is in the nature of sets that they must have either an infinite number of elements or

exactly one well-defined 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.

MATH630-1.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.

BWOC Suppose In with bn < 1 such that m n, am bn < bm is false an open interval C =

(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 l In = m*A we know by definition of infimum that a collection of open intervals

A In

{In } where l In m*A + > 0. Thus we have that O = In is an open set with m*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 1n , where O is the O associated with a

specified in the first part. Since A O > 0, we know that A G. Also since G is an

MATH630-1.nb 4

intersection of a countable collection of open intervals, G G . Finally since m*A m*O m*A +

, with 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 {In } be a countable collection of open intervals that cover A, then m*A = inf l In .

A In

Clearly {In + y} will cover A + y, with m*(A + y) = inf l In y , but l( An y ) = l(An ), so

A In

inf l In = inf l In y m*(A + y) = m*A. QED.

A In A In

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 B. Let {In } denote open

covers of A, and {Jn } denote open covers of B. Thus {In } {Jn } will cover A B. Furthermore

m*(A B) inf l In l Jn = inf l In l Jn =

A B In Jn A B In Jn

inf l In + inf l Jn = m*A + m*B = m*B.

A In B Jn

But also B A B so m*B m*(A B). Thus we must have that m*(A B) = m*B. QED.

Real Analysis - Math 630

Homework Set #2 - Chapter 3

by Bobby Rohde

9-14-00

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 1-1 operations on sets.

The last equality thus shows that E + y is measurable, QED.

MATH630-2.nb 2

Problem 10

Show that if E1 and E2 are measurable, then m(E1 E2) + m(E1 E2) =

mE1 + mE2

Proof

Since each is measurable we know that mA = m(A E1 ) + m(A E1 ), mA = m(A E2 ) + m(A

E2 ), 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 ).

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(E2 E1 )

+ m(E2 E1 ) + m(E1 E2 ) = m(E2 E1 ) + m(E2 E1 ). m(E1 E2 ) + m(E1 E2 ) = mE1 +

mE2 , QED.

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.

Example

Let En = (n, ), n . Thus mEn = n , En 1 En , En is measurable and En = .

Problem 14

a) Show that the Cantor ternary set has measure 0.

Proof

Let C0 = [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 . Thus Cn is the nth stage in the construction

of the Cantor set and Cn , n. Since Cn 1 Cn and mC0 = 1 < , by Proposition 14, we

have that m( n 0 Cn ) = limn m C n . However n 0 Cn since Cn , n 0 m( )

2 n

m( n 0 Cn ) = limn m C n = limn 3 = 0. Thus m( ) = 0. QED

MATH630-2.nb 3

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 Fn denote the construction at the nth step.

Closed

Consider Fn which must be open n. Since F = Fn , 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 F [0, 1], with a < b, such that (a, b) F = . Thus (a, b) F, with a,

b F. But this contradicts the contruction of F since every continuous interval must have its

middle removed for some Fn and (a, b) never did since the entire interval is in F. Thus a, b F

[0, 1], with a < b, c (a, b) F . Hense F is dense in [0, 1]. QED

mF = 1 -

At each phase 2n 1 intervals are removed of size 3 n. Hence mF = m([0, 1]) -

2 n

2

2n 1

m Intervals Removed = 1 - n 1 3n =1-2 n 1 3 =1- 2 * 1 32 = 1 - . mF = 1 - .

3

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 <Ei> be a sequence of

disjoint sets defined by Ei = E ri , with ri n 1 being an enumeration of the rational numbers in

[0, 1). Clearly Ei [0, 1] so m( Ei ) m[0, 1] = 1, but since Ei are pairwise disjoint, m( Ei ) =

mEi = . Which gives a contradiction. Hence mE = 0. QED

MATH630-2.nb 4

Problem 16

Show that, if A is any set with m*A > 0 then there is a non-measurable set E

A.

Proof

Let An denote A [n, n + 1), n

Since An are disjoint we know that m*A = m*( An ) = m*An n such that m*An > 0. Let P

denote the standard non-measurable set, <ri > be an enumeration of the rationals and Pi = P ri .

Let B = (An - n). Let Bi = B Pi . By problem 15, if Bi is measurable then it must must have

measure 0. Since {Bi } are pairwise disjoint and Bi = B we must have that 0 < m*B i 1 m*Bi =

0. Which is a contradiction thus a set B which is not measurable. Finally by translation

invariance of m* we know that a set An A which must be not measurable.

Real Analysis - Math 630

Homework Set #3 - Chapter 3

by Bobby Rohde

9-21-00

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 non-measurable set, and such

that f assumes each value at most once.

Counter-Example

Let E = the standard non-measurable set on [0, 1]

x, x E

Let f(x) = defined on domain [0, )

x, x E

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.

MATH630-3.nb 2

Problem 19

Let D be a dense set of real numbers. Let f be an extended real-valued

function on such that {x : f(x) > } is measurable D. Show that f is

measurable.

Proof

We wish to show that {x : f(x) > } is measurable D is equivalent to the condition {x : f(x) >

} is measurable .

1

Consider n 1 An n 1 x : f x n = {x : f(x) }. But {x : f(x) } n 1 An

since n, n < . Thus n 1 An = {x : f(x) } which means that {x : f(x) } is measurable and

hence by Proposition 18, {x : f(x) > } is measurable, so {x : f(x) > } is measurable . QED

Problem 20

Show that the sum and product of two simple functions are simple.

Proof

A B A B

1, if x A B

A B=

0, if x A B

1, if x A and x B

A B=

0, if x A or x B

But x A and x B x A B and, x A or x B x A B, thus A B A B,

QED.

MATH630-3.nb 3

A B A B A B

1, if x A B

A B=

0, if x A B

1, if x A or x B

A B A B = A B A B =

0, if x A and x B

But x A or x B x A B and, x A and x B x A B, thus

A B A B A B . QED.

A1 A

0, if x A

A=

1, if x A

1 1 0, if x A

1 A=

1 0 1, if x 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 , x D E is measurable .

A x: f x , x D E =B x: f x , x D C x: f x , x E

Clearly if B and C are measurable then A is measurable as the union of measurable sets.

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.

MATH630-3.nb 4

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 , x D is measurable .

measurable.

< 0, x : g x = x:f x , x D x :x D.

However the first term on the right is clearly measurable iff f is measurable, and the second is just x

D, but D is measurable since D is measurable. Thus x : g x is measurable for < 0 iff f is

measurable.

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 and B = x : f x are measurable for all .

Thus C = A B is measurable = x : f x .

Furthermore we know that since f is infinite on only a set of measure 0, then 0 such that mC 0 <

. Thus by Proposition 14, we have that m n 1 C 0 n = limn mC 0 n = 0. Hence by

definition of limit, D = C 0 n , for some n, with mD < / 3. QED.

MATH630-3.nb 5

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 be a simple function with values of n*( / 2) where n (-(M+1)*2/ , (M+1)*2/ ). Thus

simply choose the value for n that satisfies |f(x) - (x)| < when |f(x)| < M - - f(x) < - (x) < -f(x)

+ + f(x) > (x) > f(x) - + f(x) > n*( / 2) > f(x) - 1 + f(x) / > n / 2 > f(x) / - 1 2 +

2 * f(x) / > n > 2 * f(x) / - 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 such that it has the neccesary property. This construction will

clearly yield results of the form m M, when m f M, except possibly within / 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 = b2na , define [am, bm] = [a + (m-1)* , a + m* ], with m [0, 2n ]. Thus [am, bm]

partition [a, b] into n intervals of width .

N

We know that (x) = i 1 i * Ai , with Ai = {x : (x) = i }.

Define g x m , x am, bm , with m = i such that m(Ai [am, bm)) is the max over all i.

Clearly g(x) is a step function and in the limit n we have that g(x) = (x). Also we have that

2n

m({x : g(x) = (x)}) = m 1 m(Ai [am, bm)), for the Ai associated with m.

2n

Hence Kn = m({x : g(x) (x)}) = m 1 m(Ai [am, bm)).

However Kn 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 Kn 0 so, n such that

Kn < / 3 > 0. 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.

MATH630-3.nb 6

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 x0 < x1 < ... < xN such that for each interval (xn , xn 1 ), g(x)

assumes only one value. Let n x be any continuous function joining g(xn 6 N 1 ) and g(

xn 6 N 1 ), which does not exceed its boundary points, for n [1, N-1].

g x , if x xn 6 N 1 , n

Define h(x) =

vn 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 , which has total measure 3 NN 1 < 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

x

Let E = (- , ). Let fn x n . Thus fn x 0.

-N* ] [N* , ) is the smallest set such that x A and n N, fn x . This is true

because if we were to remove any element of A, we would add an element to A with fn x .

Hence the measure of any set, K, with the neccesary property of Proposition 23 will be larger than

mA = . Thus > 0 we have mK = > . Hence showing a counter-example.

MATH630-3.nb 7

Problem 30

Prove Ergoroff's Theorem: If < fn> is a sequence of measurable functions

that converge to a real-valued function f a.e. on a measurable set E of finite

measure, then given > 0, a subset A E with mA < such that fn

converges to f uniformly on E ~ A.

Proof

Define Ai E for i with mAi < i = 2 i ! and a smallest Ni such that x Ai , and all n Ni , |

fn x f x | < i = 1i . We know that Ai exists by Proposition 24.

Thus we have fn converging with some Nm depending only on m, when fn (x) restricted to A. So it

demonstrates uniform convergence. QED

Real Analysis - Math 630

Homework Set #4 - Chapter 4

by Bobby Rohde

9-28-00

Problem 1

a) Show that if

0, x is irrational

f x

1, x is rational

Then

b b

R a

f x x b a and R f x x 0

a

Proof

b b

R a f x x inf a x x, step functions (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 (x) f(x), x implies that

the value over any open interval of the step function (x) is 1. Thus the infimum of the integral

of all such step functions must be when (x) = 1, x.

b b

R a

f x x inf a x x b a 1 b a.

b b

R f x x sup a x x, step functions (x) f(x), x. Since irrationals are dense, each

a

interval in the composition of has an irrational in its domain. Thus (x) f(x), x that over

any open interval in , (x) 0. Thus the supremum of the integral of all such step functions must

b b

be when (x) = 0. Hence R f x x sup a x x 0. QED

a

MATH630-4.nb 2

functions such that fn increases monotonically to f. What does this imply

about changing the order of integration and the limiting process.

Construction

1 1 n 1

Let fn = x n over (0,1], so limit n we have f = 1. 0 fn x = n . So

1 1 1

limn 0 fn x 1 0 f x 0 limn fn x.

Thus we have that the limit of the integral is the integral of the limit.

MATH630-4.nb 3

Problem 2

a) Let f be a bounded function on [a, b], and let h be the upper envelope of f.

b b

Then R a

f a h.

Proof

b b

R a f x x inf a x x, step functions (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) limx y h x = inf sup h x .

00 x y

possibly where f(x) is not upper semi-continuous f(y) < limx y f x . However

hy inf sup f x so h(y) = limx y f x , when f(y) < limx y f x . Let A = {x : (x) < h(x)}

0 x y

definition of lim sup we have that > 0 such that | sup f(x) - h( ) | < / 2 and 0 < | x - | < ,

in that range such that f 2 h 2 f 2 h 2

f h f h h f . Which means that and

must be in differant intervals of the partition of , however this works > 0 which implies that

must be a boundary point between intervals of the partition. 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.

So A has at most a finite number of items in it.

b b b

Thus h except on a set of measure 0, so R a

f x x = inf a x x ah x x.

Define < n x > to be a sequence of step function such that n is over a equipartition of [a, b] into 2n

pieces, with the value of n over each piece to be sup h(x) over that piece.

BWOC Suppose limn n x h(x). That implies that > 0 such that n, | n x - h(x) |

partition intervals including x, in that interval with h( ) h(x) + . However h(x)

lim x h , which gurantees that for some small distance > 0, h(x) + > h( ) (x- , x+ ).

Thus we have a contradiction since h( ) must be < h(x) + for within of x.

Thus limn n x = h(x) and clearly n is bounded since h and f are bounded. Thus by

MATH630-4.nb 4

b b

Proposition 6 we have that ah x = limn a n x. Furthermore n are each step functions f

b b b

so that limn a n x inf a x=R a

f x x, where is any step function f.

b b b b

Hence a h x R a

f x x. So using the previous bound we have R a

f a h. QED.

integrable if and only if the set of points at which f is discontinuous has

measure 0.

Proof

b b

A bounded function f is Riemann integrable if and only if R a

f =R f.

a

b b

Since f is bounded we know R a

f a h, where h is the upper envelope from (a). By symmetry

b b

of defintion, interchanging the supremums and infimums we know that R f a g, where g is

a

the lower envelope.

b b

So f is Riemann integrable if and only if ah = a g. However 2.51(c) gives us that g(x) = h(x) if

b b b

and only f is continuous at x. So by Proposition 5(i), we know that ah = a g a h g 0.

b

But since h x g x , x we also know that h a g 0 if and only if, the set of points were

h x g x has measure 0, since otherwise some > 0 such that m( x : h x g x ) > 0,

b

and hence a h g *(measure of that set) > 0. But the set of points h x g x 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.

MATH630-4.nb 5

(iii) If f g a.e., then

Ef Eg

Hence

Ef E f

Proof

Ef Eg Ef Eg 0 E f g 0. However f g a.e. f g 0 a.e.

Thus for simple functions f g 0 a.e., sup E 0. Since 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 , so E f g 0 Ef E g. QED.

E f Ef, E f Ef E f Ef E f . QED

A mE Ef B mE

Proof

Since A f(x) B, we know from (iii) that E A Ef 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

EB B mE. Hence A m E E f B mE. QED.

A Bf Af Bf

Proof

A Bf f A B f A B , since A B = A B.

However from (i) we have that f A B = f A f B Af Bf

Real Analysis - Math 630

Homework Set #5 - Chapter 4

by Bobby Rohde

10-05-00

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.

1

Since x f x 0 n x f x 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-

1

zero. So N such that if A = x f x N , then mA > 0.

1

Define h x min f x , N , x. Thus h x f x, x. Let F A, such that 0 < mF < .

We know f sup g, and f F f , the latter coming from the fact that since f is

g f

nonnegative and F f f F with f F f, x. This implies

1

f F f sup F g F h N mF 0. Thus f 0, which contradicts the

g f

assumption that f 0, which implies that f 0 a.e. QED

MATH630-5.nb 2

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

m 1 m 1 m

2n , x n, n such that f x 2n 2n and m with m 22 n

Define n =

0, otherwise

Thus n assumes exactly 22 n -1 non-zero values over a set of measure 2n. Thus n is a

nonnegative simple function and vanishes except on a set of finite measure.

m 1

n 1 x 0, thus suppose n x = 2n for some integer m with 0 < m 22 n . This implies that

m 1 m 2 m 1 m 1 m 2m

2n f x 2n 2n 1 2n f x 2n 2n 1 , but if m 1 22 n , then clearly

2 m 1 m 1

2 m 1 2 22 n 22 n 1 , so thus we have that n 1 x 2n 1 2n n x , so its an

increasing sequence.

Let [x] denote the least natural number x, and log2 x = to the logarithim base 2 of x, with log2 0 =

- .

Given > 0 and some x0 , we may choose N = max{[ x0 ], log2 , [log2 f x0 ]}.

We wish to show that n > N, n x0 f x0 < .

Since n > x0 x0 , we have that x0 n, n so x0 meets the first condition for assigning a

non-zero value to n x0 .

m 22 n n 2 n

f x0 2n 2n 2 , for some m , m 2 . Thus x0 meets the second condition for

assigning a non-zero value to n x0 .

Finally since n > log2 ] and n meets the requirements to assign x0 a particular non-zero value,

we know that n x f x to the spacing of the values of n = 21n < 2 log

1 1

1 = . Thus

2

MATH630-5.nb 3

Proof

By definition f sup h, where h is a bounded measurable function with m x : h x 0 is

h f

finite.

We may of course limit ourselves to taking the sup over all h' x such that h' x max h x , 0 .

However all such h' x are nonnegative measurable functions. 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.

Hence we may apply part a) and conclude that h' x , an increasing sequence n such that

n h' x . Thus by Proposition 10 we have that h' lim n which equals the sup n since it

is an increasing sequence. Thus the sup over some subset of all such simple functions must be =

sup h= f . So f sup .

h f

However f, and since we are considering a sup we may assume that 0. Thus by Proposition

8 we have that f . So f sup .

MATH630-5.nb 4

Problem 5

Let f be a nonnegative integrable function. Show that the function F

defined by

x

F x f

Proof

a x

In order to show continuity we need that F a = limx a F x , x. We need f = limx a f.

1

f x, x a n

Define fn then fn is an increasing sequence with fn f as n x (- ,

0, otherwise

a).

x a 1 1

x < a, we have that f fN for some N such that x a N a N> a x .

x a a a

Thus limx a f limn fn , and by Theorem 10 we have that limn fn f . But

x a 1

also we know that x such that 0 < a - x < < 1, we have that f fM , M< , thus as

x a

0 in the limit x a , we get that limx a f limn fn .

x a a

So limx a f limn fn f.

x a x

Now consider limx a f f limx a a f , by Proposition 12. We know that since f is

integrable f < a.e. Since if f = on some set, A, of measure > 0, we would have

f mA , which is false since f , by definition of integrable. Thus without loss

of generality we may assume that f < 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

x

maximum of f y y [a, x]. Hence, a f m[a, x]*N(x), by Proposition 5. However it is clear

x

that N x1 N x2 , if a x1 x2 . Thus limx a a f limx a m a, x N x = 0.

x a x a a x

Thus limx a f f limx a a f f. Hence f = limx a f Fa =

limx a F x . F is continuous.

MATH630-5.nb 5

Problem 6

Let f n be a sequence of nonnegative measurable functions that

converge to f, and suppose fn f n. Then

f lim fn

Proof

By Fatou's Lemma we know that f lim fn . Also since fn f n, we know that fn f,

n, and thus lim fn f.

However lim fn lim fn by definition of lim sup and lim inf. So by squeezing,

lim fn f lim fn lim fn . QED

Problem 7

a) Show that we may have strict inequality in Fatou's Lemma.

Proof

Consider the sequence fn defined by fn x = 1 if n x n 1, with fn x 0 otherwise.

n 1

since n, we have fn 0 n 1 1. Thus 0 1 f lim fn . QED

b) Show that the Monotone Convergence Theorem need not hold for

decreasing sequences of functions.

Proof

Let fn x = 0 if x n and fn x = 1 for x n.

n 1

lim fn . fn n fn 1. Hence lim fn 1 0. Hence lim fn f . QED

Real Analysis - Math 630

Homework Set #6 - Chapter 4

by Bobby Rohde

10-12-00

Problem 10

a) Show that if f is integrable over E then so is f and

Ef E 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 Ef Ef . Also

Ef Ef Ef Ef E f , but since f and f are nonnegative we can drop the

asbsolute value signs, so E f Ef Ef E f . So we have shown the first part.

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 A B f f Af B f . Thus f and

f are each integrable iff f is a measurable function.

integrability of f iff f is a measurable function.

MATH630-6.nb 2

function being integrable. If f is integrable, show that the improper

Riemann integral is equal to the Lebesgue integral whenever the former

exists.

Proof

b

Let R a f x, be a Riemann integrable function with improper limit point a, and f defined for all x

(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.

Then consider the sequence of measurable integrable functions f n , such that

f x, a 1 n x b

fn x . Then lim fn x = f x a.e. Also we have that fn f . Thus by

0, otherwise

b b

Theorem 16 we have that a f lim a fn . But by construction we must have that fn x is

b b

bounded, which allows us to apply Proposition 4 to get that a fn = R a fn , and hence

b b b b b b

lim a fn lim R a fn R a lim fn R a f. So a f R a f.

Case a = -

Then consider the sequence of measurable integrable functions f n , such that

f x, n x b

fn x . Then lim fn x = f x a.e. Also we have that fn f . Thus by

0, otherwise

b b

Theorem 16 we have that a f lim a fn . But by construction we must have that fn x is

b b

bounded, which allows us to apply Proposition 4 to get that a fn = R a fn , and hence

b b b b b b

lim a fn lim R a fn R a lim fn R a f. So a f R a f. QED

MATH630-6.nb 3

Problem 11

If is a simple function, we have two definitions for ,

n

and i 1 ai mAi . Show that they are equal.

Proof

Consider , By problem 4b, we know that nonnegative measurable functions f,

f sup , with the sup taken over all simple functions 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 and are simple, we can thus deduce that

n1 n2 n1

sup i 1 bi mBi , sup i 1 ci mCi , where i 1 bi Bi and

n2 n1 n2

i 1 ci Ci . So i 1 bi mBi i 1 ci mCi , also we know that

Bi Cj i, j and each bi corresponds to some a j 0 and each ci corresponds to some

n1 n2 n1 n2 n

aj 0, so i 1 bi mBi i 1 ci mCi i 1 bi mBi i 1 ci mCi i 1 ai mAi . Since the

sets Ci and Bi , 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 fn x g x a.e. on E. Then

Proof

We know for free that lim E fn lim E fn , since this is true for any lim sups and infs.

that lim fn fn n thus By Proposition 15iii, we have that E lim fn E fn n

E lim fn lim E fn . A similar argument for lim sup fn fn gives us that lim E fn E lim fn ,

and thus we are done. QED

MATH630-6.nb 4

Problem 15

Let f be integrable over E. Then given > 0,

a) a simple function such that

E f

Construction

Since E f Ef Ef E Ef Ef E , it suffices to show that the

latter is less than . If we further specify that A x f x 0 and B x f x 0 , then

we may clearly choose = 0, x E A B . So we may further simplify the above

expression to E f 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 .

taken over all simple functions f. Since f and f are nonnegative measurable functions we

know there must exist simple functions S and I respectively such that 0 f S 2

and 0 f I 2. The second implies that 0 f I 2. Thus if we

choose over A = S and over B = I, then we have that

Ef A Ef B 2 2 . Thus we have shown the construction as

required.

E f

Construction

By the Question 3.23c, which part of the proof to Proposition 22, 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

, > 0. Let > 0 and be a step function such that E f 2. Let N be a number such

that x N, x. This is possible since simple functions are always bounded. Choose such

that = except on a set of measure 4 N with the set denoted by A, and bounded by N. Thus

E f E A f A f < E A f + A f = E A f +

A f E A f + A f E A f + A f + A

/ 2 + 2*N *mA = / 2 + 2*N* 4 N = .

Therefore E f .

MATH630-6.nb 5

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 .

Real Analysis - Math 630

Homework Set #7 - Chapter 5

by Bobby Rohde

10-19-00

Problem 1

1

Let f be the function defined by f 0 0 and f x x sin x

for x 0.

Find D f 0 , D f 0 , D f 0 , and D f 0 .

D f 0

1

f h f 0 h sin 1

D f 0 = lim h = lim h

h

= lim sin h = 1, since for arbitrarily small positive h we

h 0 h 0 h 0

1

can find a value of h which gives sine its maximal value.

D f 0

1

f h f 0 h sin 1

D f 0 = lim h = lim h

h

= lim sin h = -1, since for arbitrarily small positive h we

h 0 h 0 h 0

1

can find a value of h which gives sine its minimal value.

D f 0

1

f 0 f h h sin 1

D f 0 = lim h = lim h

h

= lim sin h = 1, since for arbitrarily small positive h

h 0 h 0 h 0

1

we can find a value of h which gives sine its maximal value.

D f 0

1

f 0 f h h sin 1

D f 0 = lim h = lim h

h

= lim sin h = -1, since for arbitrarily small positive h

h 0 h 0 h 0

1

we can find a value of h which gives sine its minimal value.

MATH630-7.nb 2

Problem 2

a) Show that D f x = D f x

Proof

f x h f x f x h f x f x f x

D f x = lim h = lim h = inf sup =

h 0 h 0 h 0 0 h

f x f x f x f x f x h f x

inf inf = sup inf = lim h = D f x . QED.

h 0 0 h h 0 0 h h 0

b) If g x = f x , then D g x = D f x

Proof

f x f x h gx gx h gx gx h gx

D f x = lim h = lim h = lim h = lim g x h h =

h 0 h 0 h 0 h 0

D g x . QED

MATH630-7.nb 3

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

f c h f c f c h f c

Well D f c = lim h D f c = lim h , since lim sups are neccesarily greater

h 0 h 0

than lim infs. Similarly D f c D f c . So we only really need to establish that

D f c 0 D f c . Since f c is a local maximum an open interval I a, b such that c

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 I.

f c h f c

Consider D f c = lim h , since c h I, f c h f c f c h f c 0.

h 0

f c h f c

Furthermore since h > 0, we know that h 0. By definition lim g h inf sup g ,

h 0 h 00 h

but as h 0 , the supremums form a nonincreasing sequence, and hence so long as we permit ±

as acceptable limits, lim g h must exist.

h 0

f c h f c f c h f c

Thus we have that lim h 0, since each h < 0, and the limit exists.

h 0

f c

f c h

Similarly D f c lim h has f c hf c h > 0 over I, and the limit must exist over the

h 0

extended reals, so lim f c hf c h 0. Hence D f c 0 D f c . QED.

h 0

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 D f a 0, and if f b is a local max then

0 D f b D f b.

MATH630-7.nb 4

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 g x such that D g x > 0, the above property holds.

Suppose x y, such that g x g y . 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 g x = lim g x hh g x < 0 since

h 0

gx h gx 0, x. Which is a contradiction, thus g has a local max.

which gives a contradiction. Thus we know that g is nondecreasing.

f x h f x gx h x h gx x gx h gx h

Therefore D f x = lim h = lim h = lim h h =

h 0 h 0 h 0

gx h gx gx h gx

lim h = lim h 0.

h 0 h 0

So if f is an arbitrary function of the type given in the problem then we know that > 0,

f x x, is a nondecreasing function. Suppose x y such that f x f y , yet we know that

f x x f y y. So this means that f x f y 0 and f x f y y x . Let =

f x f y , then > 0, and y x . However we may choose = y x 2 > 0. Which give

that *(y-x) = 2 . Which is a contradiction. Hence f x must be nondecreasing.

MATH630-7.nb 5

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.

Real Analysis - Math 630

Homework Set #8 - Chapter 5

by Bobby Rohde

11-03-00

Problem 7

a) Let f be of bounded variation (BV) on a, b . Show that for each

c 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

1

0, g x gx n

, n , 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

succifiently small that infintely many disconinuties between the current location and b. Thus we

experience arbitrarily many increases of at least 1n , which since g is monotonic implies that

gb , which is a contradiction. Hence the number is finite.

1

Let An x g x gx n , > 0}. It is clear that any discontinuity of a monotone

function must be in some An for n sufficiently large. So the set of all discontinuities is n 1 An ,

which must be countable since it is a countable union of finite sets. Thus the number of

discontinuities is countable.

MATH630-8.nb 2

Consider a fixed c. Suppose > 0 such that f over c , 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 > 0 a discontuinity in f over the region in question. We need to

show that given > 0, > 0, such that x with 0 c x 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 g x g c n1 . So we may take sufficiently small such that

all points of differance > 2 are more than away. Similarly for h. Since they contribute with

opposite sign, f x f c , were is the minimum of the two 's needed.

Hence limit x c exists. The proof however clearly proceeds likewise for x c . QED

rational point.

Construction

Let the sequence ri i 1, be an enumeration of the rationals in the interval 0, 1 . Define

r, x r

fr x .

0, x r

1

Let f x i 1 fri x 2n . Claim that f is such a function as required.

i 1 2n 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 0, 1 with x y, we have that some q x, y , since rationals are dense. So

1

f y f x fq y 2N , for some N since all terms for which fri x 0 yields fri x fri y , and

fq x 0 fq y . Thus f x is monotonicly increasing.

However since p implies that f p p h fp p p. Thus h > 0,

p p

f p h f p h 2N 0. Which implies that f p h f p h 2N , so limit h 0,

p

limx p f x limx p f x 2N . Hence f x is discontinuous at p since the limits can not be

equal.

MATH630-8.nb 3

Problem 8

a) Show that if a c b, then Tab Tac Tcb and that hence Tac Tab.

Proof

Tab Tac Tcb

By definition Tab sup t, where sup is taken over all possible partitions of a, b . However

k

t i 1 f xi f xi 1 . Then either c xi , for some i, or c xi , xi 1 , for some i.

If c xi , 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 tab is equal to tac + tcb , for those partitions.

partition t2 is to the original t t1 , since f xi 1 f xi f xi 1 f c f c f xi

f xi 1 f c f c f xi , 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 =

t2 t1 . Hence Tac Tcb Tab .

Every partition p1 , p2 over a, c and c, b respectively can be used to form a partition over a, b

by joining the two at c. However Tab = sup t over all possible partitions, so this includes the

partition formed by the union of p1 and p2 and hence Tab Tac Tcb .

Proof

Tab f g Tab f Tab g

k

Tab f g sup t f g , where t f g i 1 f g xi f g xi 1 =

k k

i 1 f xi g xi f xi 1 g xi 1 i 1 f xi f xi 1 g xi g xi 1 by the triangle

inequality, however this last term is just t f tg.

MATH630-8.nb 4

Tab c f c Tab f

k k

Tab c f sup t c f , where t c f i 1 c f xi c f xi 1 = i 1 c f xi f xi 1

k

= c i 1 f xi f xi 1 = c t f .

Problem 10

1

a) Let f be defined by f 0 0 and f x x2 sin x2

, for x 0. Is f of BV

on 1, 1 ?

Answer

No. Over any interval a, b , the diffence from x2 sin x12 , will be = a2 sin a12 b2 sin b12 .

Suppose a is a minimum, thus a12 n 2 , for some integer n. Thus an adjacent maximum at

1

b such that b2 n 2. So a2 sin a12 b2 sin b12 = n 1 1 n

1

1=

2 2

2 2

2 n 1 = 4 n28 1n . However as the number of partitions gets arbitrarily large we may

2 n 1

add arbitrarily many terms of this form with n increasing. So

0 8 n 1

T 1 f n 1 4 n2 1 n 12 n , since the last series is a constant times the

harmonic series. Therefore f is not a BV.

MATH630-8.nb 5

1

b) Let g be defined by g 0 0 and g x x2 sin 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 Tab may be broken into finitely many

1 1

pieces. We know that minima and maxima occur at sin x 1 x= n . It is obvious that

2

2

T 1< , purely by inspection and our knowledge of basic calculus. Proceed by breaking Tab into m

2

pieces such that the first piece is T 1 , and the proceeding m-2 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 n 1 , n 1 1 . So

2 2

1 2 1 2 1 2 1 2

n sin n 2 n 1 sin n 1 2 = n n 1 =

2 2 2 2

2 2

2 n 2 2 n2 14 2 n2 1

2

2

2 < n 12 n 2 2 < n 2

4

4 , note that the inequalities depend on the fact

n 2 n 1 2

that n < 0.

2 1

m 2 2 n2

Now we want to consider the total T 01 T 1+ n 1 n 2

4

4

0 , where T 0 is the part not

Ttail tail

covered in m steps. Taking limit m , the tail contribution must go to 0. So we only need that

1

2 n2 2 n2 14

the n 1 n 2

4

4 < . However this may be rewritten as n 1 2 n4

. However this is

bounded above by n32 for n large, which is a convergent sequence. Thus our sum must converge.

Hence T 01 . So T 11 x2 sin 1x is BV. QED.

MATH630-8.nb 6

Problem 14

a) Show that the sum and differance of two AC functions are also AC.

Proof

n

Let f and g be AC. Let > 0. Then 1, 2 such that i 1 f xi f xi 2 and

n

i 1 g xi g xi 2 finite collection of intervals xi, xi , such that

n

i 1 xi xi min 1 , 2 .

However in 1 f xi g xi f xi g xi n

i 1f xi f xi g xi g xi , by triangle

n

inequlaity. Further i 1 f xi f xi g xi g xi < 2 2= .

Hence we may choose min 1, 2 in order to satisfy the condition and show that f g is AC.

that -f is AC. So f g must also be AC.

Proof

2

Lemma: f is AC implies that f 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 > 0. Since f is AC, >0

n

such that i 1 f xi f xi 2 M , finite collection of intervals xi , xi , such that

n

i 1 xi xi .

n n

Consider i 1 f xi 2 f xi 2

= i 1 f xi f xi f xi f xi

n n

i 1 f xi f xi f xi f xi i 1 f xi f xi 2M < 2M 2M = .

2

Thus f is AC. Concluding the Lemma.

MATH630-8.nb 7

2 2 2

f g f g

f g 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 > 0 we may choose f constrained by 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.

1

c) If f is AC on a, b and if f is never zero there, then the function g 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

> 0. Since f is AC, > 0, such that in 1 f xi f xi m2 .

n 1 1 n f xi f xi n f xi f xi m2

Consider i 1 f xi f xi = i 1 f xi f xi i 1 m2 < m2 = .

1

Thus g f is AC. QED.

Real Analysis - Math 630

Homework Set #10 - Chapter 6

by Bobby Rohde

11-17-00

Problem 7

a) For 1 p , we denote l p the space of all sequences v v 1 such

p

that v 1 v . Prove the Minkowski ineqality for sequences

v v p v p v p.

Here we have 1 p ,

p p

v p v 1 v and v sup p

Proof

Case 1 p

1 1

p p

v p v 1 v

p

v v p = v 1 v v

p

.

If either v or v has norm 0, the proof is trivial so we may assume that each has norm greater than

1 and then define real numbers , > 0 such that v

p 1, v

p 1.

1

p

1

p

1 p p

v 1 v v

p

v 1 v v

p

= v 1

v v

=

1 1

p p p p p p p

v 1

v v

v 1

v v

,

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 = , with 1.

1

p p

So v 1 v v + = v p+ v p . QED.

MATH630-10.nb 2

Case p

v v sup v p sup v v sup v + sup v =

v v . QED.

1 1

b) Show that if v l p and v l q with p q

1, then

v 1 v v v p v q

Proof

Case p 1, q

v 1 v v v 1 v sup v = v 1 v v = v 1 v .

Case 1 p

q

Define v p . We may chosoe to assume that v and vare positive since the norm is dependant

p 1 p p

only on their absolute value. By Lemma 3 p t v =p t v t v v

p p

Summing both sides. v 1 p t v v v 1 v t v v =

p- p p p

v t v p v p v p t v p v p

v 1 v v p p

v p v q . QED.

Problem 10

Let f n be a sequence of functions in L . 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.

Proof

fn f in L imples f n f 0. Thus 0 N such that n N, f n f .

except on a set E of measure 0, fn f . Thus fn is uniformly converges to f on E.

and n N, fn x f x . except on E, f x < fn x ess sup fn x . Thus f is

bounded a.e. and hence f is in L .

MATH630-10.nb 3

Problem 11

Prove that L is complete.

Proof

In order to show this we need that every Cauchy sequence of functions fn in L converges to

some f in L .

norm we know that fn fm is bounded a.e. by .

Define the pointwise limit f x .

, otherwise

It suffices to show that ess sup f , since if fn converges then it must converge to f a.e.

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 fn x f x if limn fn x exists. Since fn is in L it

is bounded by some ess sup . Then where the limit exists, f is bounded by + .

Thus we only need to show that the limit must exist upto a set of measure 0.

MATH630-10.nb 4

Problem 16

Let f n be a sequence of functions in L p, 1 p , which converge a.e.

p

to a function f in L . Show that f n converges to f in L p if and only if

fn f .

Proof

If fn f fn f 0

1 1 1

p p p p p p

fn p

f p

fn f p

0, since p 1. However fn f

p

fn f , for p 1

1

p p

fn converges to f in L p fn f 0 fn f p

0. However, fn f

fn f p.

??????????????????????????????????

Show that f , g L , such that step functions and continuous

functions , f 0 and g 0.

Counter-example on step functions.

sin 1x , x 0

Define f x .

0, x 0

0 0 1 2 ... N 1 . Let denote the size of the smallest interval in the partition.

Consider intervals of the form An = n1 , n 11 , for n . Clearly An 0, 1 , and infinitely

many An of length < . So choose n, m such that An m , m 1 and length An . Clearly by

1

construction f An 0, 1 , so regardless of the value of over m, m 1 , f 2 over An .

1

Hence f 2 . Thus showing the counter example to Proposition 8.

MATH630-10.nb 5

1

0, x 2

Define g x 1

.

1, x 2

1

Clearly g is in L . Consider a continuous function approximating g. By continuity at x 2 ,

>0 > 0 such that x 12 , x 1

2 .

2 and the interval 1

2 , 1

2 where x 1

2 , yet g = 1

1

over this interval so g 2.

1 1

1 1 1

Suppose that 2 0, then choose = 2

2 and the interval 2 , 2 where x 2

2 .

1

So g 2

2 . Thus g > 0. Thus showing the counter example for Proposition

8 at p = . QED.

Real Analysis - Math 630

Homework Set #11 - Chapter 6 & 11

by Bobby Rohde

11-25-00

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

, 0

Let f x sgn g x , where sgn . Since g 1 0, that implies that

0, 0

m x:g x 0 0 ess sup f(x) = 1.

Case g 1 0

Let f x 0, x. Then f . f g = g g 0 finite . Thus

f g= g 1 f = 0. QED.

MATH630-11.nb 2

integrable function f such that

f g g f 1

Proof

By definition ess sup g(x) = inf M : m t : f t M 0. Given > 0. Let

E x:g x g . Thus mE > 0. Let f x E. Then f g = E g

E g by definition of E. Hence we have this = mE g =

g f 1 . Thus f g g f 1 . QED.

Problem 6.22

Find a representation for the bounded linear functionals on l p, 1 p .

Construction

Problem 6.24

Show that the element g in L p given by Theorem 13 is unique.

Proof

MATH630-11.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. 0.

ii. E E 0 E

iii. E 0 E A B where B and A C, C , C 0.

Proof

First we must show that the 0 defined by (iii) is in fact a -algebra. Thus we need that

E1 , E2 0, E1 E2 0 and E1 0 and the union of any countable collection of sets in 0

is in 0 .

E2 . Thus E1 E2 = A1 A2 B1 B2 . And we know that B1 B2 , since is a -

algebra. Furthermore A1 A2 C1 C2 . C1 C2 , since is a -algebra and

0 C1 C2 C1 C2 0. So C1 C2 0.

E1 = A1 B1 = A1 B1 = C1 A1 B1 B1 C1 . However B1 C1 , and

hence B1 C1 . Also C1 A1 B1 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

n 1 En , En 0 An Bn = Bn An , we know that Bn , so we only need

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.

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.

MATH630-11.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.

Problem 11.10

Prove Proposition 7: Let f be a nonnegative measurable function . Then

a sequence n of simple functions with n 1 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

22 n

Define En,k x:k 2 n f x k 1 2 n and n 2 n

k 0k En,k , with respect to each

pair of integers n, k .

2 n

Since f is finite, for a given n, k , n 0, k 0, mEn,k . Thus m k2 1 En,k , and so 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 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.

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 .

MATH630-11.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 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 f inf , then clearly a sequence of monotone increasing simple functions

f

n which converge to g which will be less than f a.e. By montone convergence theorem

g lim n inf f . Hence f g. QED.

f

Real Analysis - Math 630

Homework Set #12 - Chapter 12

by Bobby Rohde

11-25-00

Problem 12.2

Assume that Ei is a sequence of disjoint measurable sets and E = Ei .

Then A we have that

A E A Ei

Proof

If Ei has only one non-empty set element then this is trivial. Suppose Ei has a finite

number of non-empty sets, then by the assumption of the induction hypothesis say that

n 1 n 1

E i 1 Ei , conforms to the property that A E i 1 A Ei .

n

Consider E i 1 Ei, then A E = A E En A E En , since En is

measurable. However E En , and exploitning the fact that each Ei is disjoint, we arrive at

A E = A En A E = in 1 A Ei . Thus the induction is proved. Hence we

know for any finite sequence this property will hold.

N N N

So N, A i 1 Ei i 1 A Ei . However A i 1 Ei A i 1 Ei , so

N

A E A i 1 Ei i 1 A Ei . The left hand side is now independant of N so

we make take the right hand side to infinity. So A E i 1 A Ei , yet we know form

the properties of outer measure that A E i 1 A Ei . Hence

A E A Ei . QED.

MATH630-12.nb 2

Problem 12.4

Prove Proposition 9 - Let be a semialgebra of sets and a nonnegative set

function defined on with 0 if . 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 Ci of sets in , then

C Ci

C Ci

a) Condition (i) implies that if A is the union of each of two finite disjoint

collections Ci and Di of sets in , then Ci D j.

Proof

A Ci by condition (i), but also by condition (i) we have that A D j , so be the

transitivity of equality, Ci D j.

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 Ci , for any disjoint sequence

Ci , such that Ci 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.

MATH630-12.nb 3

Problem 12.5

Let be a semialgebra of sets and the smallest algebra of sets containing

.

n

a) Show that is comprised of sets of the form A = i 1 Ci with Ci .

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.

n m

Let A, B , such that A i 1 Ai , and B i 1 Bi , where each set Ai and Bi is in . Then

n m n m n m

A B i 1 Ai i 1 Bi = i 1 Ai j 1 Bj = i 1 j 1 Ai B j , but each Ai B j ,

since is closed under finite intersections. Thus A B is a finite union and hence an element of

of the appropriate form. By induction is closed under finite intersections.

ni

Consider A = ni 1 Ci . However by definition of semialgebra, Ci k 1 Di,k , with Di,k . Thus

n ni ni

A i 1 k 1 Di,k , but k 1 Di,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 in 1 Ci with Ci , yet no other

elements have been added besides these.

by and , respectively.

Proof

, thus . However contains all elements of the form i 1 Ci , where Ci .

n

Hence for each element in , A = i 1 Ci ,

A i 1 Ci . Thus = . Thus

= . QED.

MATH630-12.nb 4

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 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 with B , C . Then we want to show that B C . We know that there must

exist a sequence of sets Bi , with each set in such that Bi B, since it appears in the

closure of with respect to unions.

Define Ei to be the sequence of sets such that Ei Bi C , since is closed under finite

intersections. However the Ei Bi C B C. Thus the intersection of any element of

with one of is in . By induction we may intersect any finite number of elements in

with elements of .

the elements of a generating sequence for C. This is equivalent to B Ci , however we know

that B Ci from above, and thus taking the countable union will also be in . is

closed under finite intersections.

sets in

Construction

Clearly if the set in question is a member of then the result is trivial, since we just take every

term in the sequence to be that set.

i 1

Define Ai j i Bj k 0 Bk , with A0 Bi . Given any element x B, x Bk , k,

hence x Ai , i. Given y B, y BN for some N and hence y Ai , i N. If N is the least

such N that works then y Ai , i N, and thus Ai 's form a decreasing sequence and Ai B.

Thus we have a countable intersection of decreasing sets in A that yields B A .

MATH630-12.nb 5

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 E A .

Proof

Real Analysis - Math 630

Homework Set #13 - Chapter 12

by Bobby Rohde

12-11-00

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.

First considering the Tonelli theorem, we know that the set of positive integers is -finite, so we

may apply the Tonelli theorem provided that f is nonnegative everywhere.

cardinality 0 we know that an enumeration of the points in X Y. Let ri i 1 be such an

enumeration. Then we can equate f x, y = i 1 f ri ri x, y . If we further define

n

n x, y i 1 f ri ri 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 n , and by

application of the monotone convergence theorem, f limn n . However the right hand

n

side is equivalent to limn i 1 f ri ri = i 1 f ri , since the counting measure over a

set of one element is 1. Thus saying that X Y f is equivalent to asking that

i 1 f ri and i 1 f ri .

MATH630-13.nb 2

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.

Y.

Well fx y i 1 f ri ri x, y , where x is fixed. Which is equal to i 1 f x, i i y.

Consider y : fx y = A . We need only show that A is well defined for all . However A

is some subset of Y, and thus A Y , so A is measurable, since it is in the collection of

measurable sets, and hence fx y is a measurable function by Proposition 11.5.

Let us explicitly compute Y f x, y y = Y fx y = Y i 1 f x, i i y . 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 fx y = i 1 f x, i i = i 1 f x, i , since i 1, i.

That the function i 1 f x, i is measurable follows immediately form the fact that x X , and =

(X). The argument is the same as the 2nd paragraph in (i).

(iii) X Y f x, y X Y f x, y

Working from the conclusion of (ii), that Y fx y i 1 f x, i , we will then integrate with

respect to , to find that X Y f x, y = i 1 j i f i, j i . (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 sum, and the integrals are thus the same by Monotone

convergence theorem.) However i 1 since is the counting measure, and thus

X Y f x, y = i 1 j i f i, j . However we also know that X Y f x, y =

X Y i 1 f ri ri x, y = i 1 f ri ri = i 1 f ri , since

ri 1, i . However since the ri 's range over ever pair j, k of integers exactly once,

this is identical to the above result.

Hence X Y f x, y X Y f x, y = i 1 j i f i, j .

MATH630-13.nb 3

Problem 12.22

Let h and g be integrable functions on X and Y, and define

f x, y h x g y . Then f is integrable on X Y and

X Y f Xh Yg

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.

n n

inf X Y i 1 ci Ei X Y , where x, y i 1 ci Ei x, y is a

f

nonnegative simple function and each Ei X Y.

m1 m2

Xh inf X h i 1 di Hi X and Yg inf Y g i 1 fi Gi Y , with

h h g g

m1 m2

h x i 1 di Hi x and g y i 1 fi Gi y.

n m1 m2 m1 m2

i 1 ci Ei x, y i 1 di Hi x i 1 fi Gi y = i 1 j 1 di f j Hi x G j y . However

Hi x G j y = Hi G j x, y .

Ei H j i Gk i and ci d j i fk i , where the j, k depend on i and each pair of j, k are used

exactly once over the m1 m2 terms.

that each G j and Hk is in X and Y respectively since these are the basis sets, so this devolves to

m1 m2

i 1 dj i fk i H j i Gk i . However H j Gk is a rectangle in X Y, so

H j i Gk i = H j i Gk i

m1 m2 m1 m2

i 1 dj i

fk i H j i Gk i = i 1 dj i fk i Hj i Gk i , but by definition of

the j i and k i this is just equal to im11 di Hi m2

i 1 fi Gi = X h Y g .

MATH630-13.nb 4

f h h g g

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 Yf Xh Yg , if the integral over f exists.

f f

inequality is always true of the sups and infs used to define integrals.

f f

exists and its value is Xh Yg . 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.

MATH630-13.nb 5

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

2 2 x, if x y

f 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 x, y 2 2 x x, y , where

A B

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 Y X f x, y .

2 2 x Ax 2 2 x Bx . Furthermore we know that for a fixed x, Ax has only one

point determined by x y, and similarly Bx has only one point. Hence the counting measure

evaluated on these sets is just 1. Thus Y f x, y = 2 2 x 2 2 x 0. Hence

X Y f x, y 0.

gives the integral as 2 2 y Ay 2 2 y 1 B 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 2 2 y 2 2 y 1 = 2 y 12 1 2 y 1 . Also at y 1,

3

X f x, y = 2 2 1 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 =

3 y 1 3 N i 1

1,2,... ,N 2 1 2 1 = 2 1 i 2 2 i Since we are using the

counting measure, each evaluation is in fact one. Simplifying the geometric series we have

MATH630-13.nb 6

1

3 1 1

2 1

2

1 = 2 . Thus Y X f x, y 2 0 X Y f x, y . Hence neither the

2

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 and are -finite from

the Tonelli Theorem: Let X Y be the interval 0, 1 , with the class

of Borel sets. Let be the Lebesgue measure and the the counting

measure. Then the diagonal x, y X Y : x y is measurable

(is an , 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.

This means that Y x, y = x , but x

y is only non-zero for the one value where x y,

and so since is the counting measure, we have that Y x, y = 1. So X Y x, y =

X1 = X 1.

Which evaluates the integral to y , which again is a set containing only the point where x y.

However since is the Lesbegue measure this evaluates to 0. So we have that Y X x, y

= Y0 0. Since 0 1, we have thus shown that fails to satisfy the (iii) property of the

Fubini and Tunelli Theorems. QED.

MATH630-13.nb 7

Problem 12.31

Let f be a nonnegative measurable function on , , and let m2 be the

two-dimensional Lebesgue measure on 2. Then

m2 x, y :0 y f x m2 x, y :0 y f x f x x

0 t t f x x.

Proof

m2 x, y :0 y f x m2 x, y :0 y f x f x x

Essentially in the first part we are attempting to show that f x x is the same thing as the area

under the curve (with and without its boundary).

measure. Consider A , since is -finite and is everywhere nonnegative, we may

apply Tonelli Theorem to conclude that A A Ax . For any

particular x we have that Ax is the interval 0, f x , and thus that Ax f x . Thus A

= f x , but the integral with respect to Lesbegue measure is exactly what is intended by the

notation f x x. Finally we know that A = A by the definition of an integral

of a characteristic function, but m2 . Hence we have that m2 A = f x x.

If we then note that B behaves exactly the same as A in the above proof since Bx = 0, f x , and

hence Bx 0, f x = f x . Thus the conclusion is equally valid that m2 B f x x.

MATH630-13.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 , 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 is a decreasing function, this follows immediately from that fact that

mA mB, whenever A ! B, and the observation that if x x : f x t1 then x x : f x t2 ,

t2 t1 . Thus for t1 t2 , we have that x : f x t1 " x : f x t2 . So m x : f x t1

m x : f x t2 .

Let x, t :f x t 0

equality coming from the fact that the measure of a cross-section is the same as the integral of its

characteristic function. Since characteristic functions are bounded and both measure are taken with

respect to the -finite real numbers, we may apply Tonelli thoerem to conclude that

0 x, t x t = 0 x, t t x. From here we need only observe that is the same

as A in the first section of this problem, upto the labelling of variables. Thus from above we may

conclude that 0 x, t t x = f x x. 0 t t f x x. QED.