Completeness of \ell^1: Proving Convergence of Cauchy Sequences

  • Thread starter Thread starter winter85
  • Start date Start date
  • Tags Tags
    Complete
Click For Summary

Homework Help Overview

The discussion revolves around proving the completeness of the space \(\ell^1\), which consists of all real sequences whose absolute sums converge. The original poster is attempting to show that a limit sequence derived from a Cauchy sequence of sequences in \(\ell^1\) also belongs to \(\ell^1\).

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning

Approaches and Questions Raised

  • The original poster discusses defining a limit sequence \(u\) from the Cauchy sequence \(v_n\) and questions how to prove that \(u\) converges absolutely. They express uncertainty about handling the sum of terms when considering convergence.
  • Another participant suggests a method for justifying the interchange of the sum and limit, outlining a series of steps to establish the necessary bounds.

Discussion Status

The discussion is active, with participants exploring the conditions under which the limit sequence converges absolutely. Some guidance has been offered regarding the interchange of summation and limits, and the original poster expresses clarity following this input.

Contextual Notes

Participants are working within the constraints of proving properties of sequences in a normed space, specifically focusing on the implications of Cauchy sequences and absolute convergence.

winter85
Messages
33
Reaction score
0

Homework Statement



Prove that [tex]\ell^1[/tex], the space of all (real) sequences [tex]v = \{v_k\}[/tex] such that [tex]\sum|x_k| < \infty[/tex], is complete.

Homework Equations



[tex]\ell^1[/tex] is a normed space with the norm [tex]||x|| = \sum |x_k|[/tex]

The Attempt at a Solution



Let [tex]v_n[/tex] be a Cauchy sequence of sequences in [tex]\ell^1[/tex]. Then for all [tex]\epsilon > 0[/tex] there exists N > 0 such that for all n,m > N we have [tex]\sum |v_{n,k} - v_{m,k}| < \epsilon[/tex] (here [tex]v_{n,k}[/tex] means the kth term of the nth sequence)

in particular this means that [tex]|v_{n,k} - v_{m,k}| < \epsilon[/tex] so we can define a sequence [tex]u = \{u_k\}[/tex] as [tex]u_k = \lim v_{n,k}[/tex] as n goes to infinity.

Now i think the sequence u would be the limit of [tex]v_n[/tex] as n goes to inifnity, but I'm not sure how to prove it. Firstly, I don't know how to prove that u converges absolutely. the problem is by the definition of u, given [tex]\epsilon[/tex] I can find a sequence v_n whose terms are each within [tex]\epsilon[\tex] from the corresponding term in u, but when summing, this is like summin [tex]\epsilon[\tex] infinitly many times.. so how can I do it? any hint would be appreciated :)<br /> Thanks.[/tex][/tex]
 
Physics news on Phys.org
if we do this:
[tex]\lim \sum |v_{n,k} - v_{m,k}| = \sum \lim |v_{n,k} - v_{m,k}|[/tex] where the limit is taken as m -> infinity, then we have [tex]\sum |v_{n,k} - u_{k}| < \epsilon[/tex] for sufficiently large n. But how can I justify interchanging the sum and the limit?
 
Your u is good.

There is a trick to interchanging the sum and the limit, which roughly speaking is this:

(1) use epsilon,
(2) change the bound on the infinite sum to a bound on a finite sum from k=1 to j,
(3) note that the bound is independent of j,
(4) take your limit as m approaches infinity (OK to do, since sum is finite),
(5) bound is still independent of j, so now let j approach infinity.

Here are the details, applied to the problem of showing u converges absolutely.

(1) [tex]v_n[/tex] is Cauchy so it's bounded; thus [tex]||v_n||\le M[/tex] (using M instead of epsilon for this example)

(2) for any j, [tex]\sum_{k=1}^j |v_{n,k}|\le M[/tex]

(3) previous bound is independent of j

(4) take limit as n (using n instead of m) approaches infinity, [tex]\sum_{k=1}^j |u_k|\le M[/tex]

(5) bound is still independent of j, so let j approach infinity, [tex]||u||\le M[/tex].

Now you can try it for the convergence.
 
oh, I see! It's now very clear to me. Thank you so much. :)
 

Similar threads

  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
3
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 37 ·
2
Replies
37
Views
6K
Replies
4
Views
2K
  • · Replies 17 ·
Replies
17
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K