Is the order of limits interchangeable?

  • Context: Graduate 
  • Thread starter Thread starter Jhenrique
  • Start date Start date
  • Tags Tags
    Limits
Click For Summary

Discussion Overview

The discussion revolves around the interchangeability of limits, specifically whether the order of limits can be switched in mathematical expressions. This includes theoretical considerations and counterexamples related to real analysis.

Discussion Character

  • Debate/contested
  • Technical explanation
  • Mathematical reasoning

Main Points Raised

  • One participant asserts that the identity $$\lim_A \lim_B = \lim_B \lim_A$$ is generally true.
  • Another participant disagrees, providing a counterexample using a specific function defined on natural numbers, demonstrating that the limits can yield different results depending on the order.
  • A later reply questions whether there are general cases where the interchangeability of limits holds true.
  • Another participant confirms that there are indeed general cases where the limits can be interchanged, referencing significant theorems in real analysis that address this issue.
  • Specific theorems such as the Monotone Convergence Theorem, Dominated Convergence Theorem, and Fubini's Theorem are mentioned as relevant to the discussion.

Areas of Agreement / Disagreement

Participants do not reach a consensus; there are competing views regarding the interchangeability of limits, with some asserting it is not generally true while others argue that there are conditions under which it is valid.

Contextual Notes

The discussion highlights the complexity of limit interchangeability, with references to specific mathematical contexts and theorems that may apply, but does not resolve the conditions under which limits can be interchanged.

Jhenrique
Messages
676
Reaction score
4
In the sense most ample and general of limits, the following identitie is true:
$$\\ \lim_A \lim_B = \lim_B \lim_A$$
?
 
Last edited:
Physics news on Phys.org
No, it's not.
 
  • Like
Likes   Reactions: 1 person
Simple counterexample: consider the function ##x : \mathbb{N} \times \mathbb{N} \rightarrow \mathbb{N}## defined by
$$x(m,n) =
\begin{cases}
1 & \text{if }m > n \\
0 & \text{otherwise}
\end{cases}$$
For every ##m##, we have ##\lim_{n \rightarrow \infty}x(m,n) = 0## and therefore ##\lim_{m \rightarrow \infty}\lim_{n \rightarrow \infty}x(m,n) = 0##.

Similarly, for every ##n##, we have ##\lim_{m \rightarrow \infty}x(m,n) = 1##, and therefore ##\lim_{n \rightarrow \infty} \lim_{m \rightarrow \infty}x(m,n) = 1##.
 
and exist general cases where ##\\ \lim_A \lim_B = \lim_B \lim_A## is true?
 
Jhenrique said:
and exist general cases where ##\\ \lim_A \lim_B = \lim_B \lim_A## is true?

Yes, and that's actually what a giant part of real analysis is about: finding when you can switch two limits.

Please see Knapp's "Basic Real Analysis". In the first chapter he already gives ##2## general situations where it's true.
Aside from that, there are many specialized situations where it is also true, these are incredibly important theorems. A small selection:
http://en.wikipedia.org/wiki/Monotone_convergence_theorem#Lebesgue.27s_monotone_convergence_theorem
http://en.wikipedia.org/wiki/Dominated_convergence_theorem
http://en.wikipedia.org/wiki/Fubini's_theorem
http://en.wikipedia.org/wiki/Power_series#Differentiation_and_integration
 
  • Like
Likes   Reactions: 1 person

Similar threads

  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 8 ·
Replies
8
Views
5K
  • · Replies 20 ·
Replies
20
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 16 ·
Replies
16
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 1 ·
Replies
1
Views
1K