Help for test - functional analysis

Click For Summary

Discussion Overview

The discussion revolves around a set of functional analysis problems posted by a student seeking assistance with solutions. The problems involve concepts such as sigma additive measures, properties of a specific transformation function, and the behavior of linear operators defined on spaces of infinite series. Participants engage in clarifying definitions, correcting errors, and exploring the implications of the problems presented.

Discussion Character

  • Homework-related
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant questions the definition of a measurable set and suggests that the conclusion regarding the sigma additive measure may not be correctly stated.
  • Another participant provides a proof for the inequality involving the transformation function, using the mean value theorem, and asserts that no constant exists that satisfies the second part of the question.
  • There is a discussion about the definition of the operator L, with one participant asserting that it is a linear operator and providing a calculation of its norm, while another participant challenges the applicability of L to certain spaces.
  • A participant expresses confusion regarding the second part of the third question, suggesting that the finite nature of non-zero elements should lead to a finite norm.
  • Another participant clarifies their earlier definition of a sequence and discusses the implications of this on the transformation properties of L, concluding that L cannot be a transformation from one space to another due to divergence issues.

Areas of Agreement / Disagreement

Participants exhibit a mix of agreement and disagreement, particularly regarding the definitions and properties of the operator L. While some points are clarified, multiple competing views remain on the implications of the problems and the correctness of certain definitions.

Contextual Notes

There are unresolved issues regarding the definitions of norms and the behavior of the operator L, particularly in relation to the spaces involved. Some assumptions about the nature of sequences and their membership in defined spaces are also under discussion.

simpleton1
Messages
14
Reaction score
0
Hi - my professor in functional analysis posted 4 prior years tests just 4 days before the test without solutions.
I'd appreciate if anyone can help send solutions for the following with the following questions :

1. $\mu$ is a sigma additive measure over sigma algebra $\Sigma$.
A $\in \varSigma$ is some finite set and we'll define the function m:$\Sigma$ to R
such that $m(B) = \mu (A \cup B)$.
Show that m is a sigma additive measure iff $\mu (A) = 0$.

2. Let f:[0,$\infty$) to [0,$\infty$) transformation given by f(x)=x+exp(-x)
Prove that $\left| f(x)-f(y) \right| < \left| x-y \right|$ for every $x\ne y$.
Is there a constant a between 0 and 1 such that $\left| f(x)-f(y) \right| < a\left| x-y \right|$
for every $x\ne y$ at [0,$\infty$)

3. X is the sapce of all infinite series of real numbers with a finite number of
elements which are different than 0 (mark elements by (a1,a2,...)).
Define two norms over X :
$\left\lVert{a}\right\rVert{}_1{}$ = $\sum_{n}^{\infty} \left| a{}_n{} \right|$
and
$\left\lVert{a}\right\rVert{}_\infty{}$ = $max( \left| a{}_n{} \right| )$

Define operator L so that L (from X to X) shifts elements left and divides by their location :
(a1,a2,...) change to (a2/1,a2/3,a3/4,...).
Is L an obstructed linear transformation from X1 to X1? If yes - what is it's Norm?
Is L and obstrcuted linear transofrmation from $\left\lVert{X}\right\rVert{}_\infty{}$ to $\left\lVert{X}\right\rVert{}_1$
 
Last edited:
Physics news on Phys.org
Re: help for test - functional analysis\mu

Hello again, simpleton,

Please edit your post, as it contains many errors.

1. The term "group" should be replaced with "measurable set".

2. The conclusion does not make sense. Did you mean "There exists a constant $\alpha$ such that for all $x,y\in [0,\infty)$, $x\neq y$ implies $|f(x) - f(y)| < \alpha|x - y|$?"

3. You've defined $\|a\|_1$ and $\|a\|_\infty$ exactly the same. It should be $\|a\|_\infty = \sup\{|a_n| :n\in \Bbb N\}$. The definition of $L$ seems off -- shouldn't it be $L(a) = \left(\frac{a_2}{1}, \frac{a_3}{2}, \frac{a_4}{3},\ldots\right)$?

Also, please put some space between problems so that it's easier to read.
 
Re: help for test - functional analysis\mu

Hi Euge - thanks for the commentation. I did post the questions under the influence of tiredness. :D

I've fixed most things I think. You are right about the definition of L in question 3. I just got too lazy to
Latex it.
 
1. It follows from the fact that $m(\emptyset) = 0$ if and only if $\mu(A) = 0$.

2. Note that $f'(x) = 1 - \exp(-x) < 1$ for all $x > 0$. Now use the mean value theorem to establish the inequality $\lvert f(x) - f(y)\rvert < \lvert x - y\rvert$ for all $x,y\in [0,\infty)$ with $x\neq y$. For the second statement, the answer is no. For otherwise $f'$ is bounded by $\alpha$ on $[0,\infty)$. Then $\sup\{f'(x) : x\in [0,\infty)\} = 1 > \alpha$.

3. If my definition of $L$ is correct (which still differs from yours), then $L$ is a linear operator on $X$. For all $a\in X_1$,

$$\|L(a)\|_1 = \sum_{n = 2}^\infty \frac{\lvert a_n\rvert}{n-1} \le \sum_{n = 2}^\infty \lvert a_n\rvert \le \|a\|_1,$$

thus $\|L\|_{X_1\to X_1} \le 1$. On the other hand, $\|(0,1,0,0,\ldots)\|_1 = 1$ and $\|L(0,1,0,0,\ldots)\|_1 = \|(1,0,0,0,\ldots)\|_1 = 1$. Therefore, $\|L\|_{X_1\to X_1} \ge 1$, proving that $\|L\|_{X_1\to X_1} = 1$.

However, $L$ is not a transformation from $X_\infty$ to $X_1$, since $a = (1,1,1,1,\ldots) \in X_\infty$ (with $\|a\|_\infty = 1$), but $\|L(a)\|_1 = \sum\limits_{n = 1}^\infty \frac{1}{n} = +\infty$.
 
Ok, Thanks.
I got the first two questions and the first part of question 3.
But the second part of 3 has still got me confused - since the number of elements in a different
than 0 are finite , the sum of $\left\lVert{a}\right\rVert\infty$ over a=(1,1,1,1,...0,0,0) will not
be $\infty$ but a finite number.
 
I was going to correct the error but you beat me to it! Actually, I defined $a$ to be the sequence of $1$'s, no zeros are in the sequence. That is a problem since then $a$ does not belong to $X$.

Let's try this again...

$L$ is not a transformation from $X_\infty$ to $X_1$. For given $m\in \Bbb N$, let $a^m = (1,1,\ldots,\underbrace{1}_{m+1},0,0,\ldots)$. Then $\|a^m\|_\infty = 1$ and $\|L(a^m)\|_1 = H_m$, then $m$th harmonic sum. Hence $\|L\|_{X_\infty\to X_1} \ge H_m$. Since $m$ was arbitrary and $(H_m)_{m\in \Bbb N}$ is a divergent sequence, then $\|L\|_{X_\infty\to X_1} = \infty$.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
21
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K