Is the Space of Absolutely Continuous Functions Complete?

Click For Summary
The discussion centers around the completeness of the space of absolutely continuous functions, with participants questioning the properties of these functions and their norms. It is established that the completeness depends on the chosen norm; for instance, using the norm ||f||=max|f| may lead to incompleteness, while ||f||=max|f|+max|f'| ensures completeness. The conversation also highlights confusion regarding the relationship between absolutely continuous functions and their derivatives, noting that these functions do not necessarily possess k-th derivatives. A key point raised is that to discuss completeness, one must specify the norm being used. The thread concludes with a clarification on the implications of convergence in normed spaces, emphasizing that an absolutely convergent series indicates completeness.
Matthollyw00d
Messages
92
Reaction score
0
Is the space of all absolutely continuous functions complete?
I've never learned about absolutely continuous functions, and so I'm unsure of their properties when working with them. I'm fairly certain it is, but would like some verification.
Or a link to something on them besides the wikipedia page could be useful as well.

Much appreciated.
 
Physics news on Phys.org
It depends on what you use for the norm. For example (for simplicity I'll assume a finite domain) - if ||f||=max|f|, the answer would be no. On the other hand if ||f||=max|f|+max|f'|, the answer would be yes.
 
I was thinking with the induced norm from <f,g>=\Sigma_{k=1}^n \int^1_0 f^{(k)}(t)\overline{g^{(k)}(t)}dt
 
I'm confused? How are f(k) related to f?
 
The kth derivatives.
I'm dealing with a subset of absolute continuous functions and trying to show completeness. It seemed easier to just show it was closed if I knew the space of all absolutely continuous functions were complete w.r.t the induced metric from the above inner product. But now I don't think we can talk about that space w.r.t the inner product, since absolute continuity gives nothing about the kth derivatives existing.

I guess I'm left with showing that an arbitrary Cauchy sequence converges.
 
What is n? And indeed, absolutely continuous functions need not have k-th derivatives (for any k), so this inner product does not make any sense.

So back to start: what norm do you want to use (you can't talk about completeness without talking about a norm)?
 
Well I was dealing with a subset where it did make sense.

Let me just ask a different question:
If \{h_n\} is a sequence in H and \Sigma^\infty_{n=1}||h_n||<\infty and I show that \Sigma^\infty_{n=1}h_n converges in H; does that imply that H is complete?
 
Assuming that (h_n) is an arbitrary sequence: yes, one of the equivalent formulations of completeness of a normed space is "every absolutely convergent series converges".

\\edit: for a proof of this, see e.g. here.
 
Last edited:

Similar threads

  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 16 ·
Replies
16
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
3
Views
3K
Replies
7
Views
1K
  • · Replies 3 ·
Replies
3
Views
23K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K