Absolute convergence Text Book question: Boas 3rd Ed

Click For Summary

Discussion Overview

The discussion revolves around the concept of absolute convergence as presented in Mary L. Boas' Mathematical Methods in the Physical Sciences. Participants explore the definition of convergence for alternating series, particularly focusing on the treatment of limits and absolute values in the context of series convergence tests.

Discussion Character

  • Technical explanation
  • Conceptual clarification
  • Debate/contested

Main Points Raised

  • One participant questions the definition of absolute convergence in the textbook, specifically why the limit is stated without absolute values, suggesting potential confusion about whether the negative sign in alternating series should be disregarded.
  • Another participant asserts that the limit of a sequence converging to zero is equivalent whether or not absolute values are considered, implying that the distinction may not be significant.
  • A participant reflects on their experience with power series and how recognizing that the limit of the absolute value can simplify convergence tests, particularly when testing boundary points.
  • One participant raises a related question about the integral test, asking if the alternating term can be ignored when determining convergence, and expresses uncertainty about the implications of integrating without the alternating factor.
  • Another participant confirms that an infinite integral can be treated as a limit, but does not resolve the question regarding the integral test for alternating series.

Areas of Agreement / Disagreement

Participants express differing views on the treatment of absolute values in limits and the implications for convergence tests. There is no consensus on whether the alternating term can be disregarded in the context of the integral test.

Contextual Notes

Participants highlight potential misunderstandings regarding the definitions and applications of convergence tests, particularly in relation to limits and the treatment of alternating series. The discussion reflects varying levels of familiarity with the concepts involved.

Who May Find This Useful

This discussion may be useful for students and educators in mathematics and physics who are grappling with the concepts of convergence in series, particularly in the context of alternating series and integral tests.

Battlemage!
Messages
292
Reaction score
44
In Mary L. Boas' Mathematical Methods in the Physical Science, 3rd ed, on page 17 it goes over absolute convergence, and defines the test for alternating series as follows:

An alternating series converges if the absolute value of the terms decreases steadily to zero, that is, if |an+1| ≤ |an|, and limn→∞ an = 0.​

Now, my confusion starts with the limit. In the textbook, the limit is NOT of the absolute value of an, otherwise they would have written |an|, right?

In the example given to show this definition in operation, they are using the series ∑ [(-1)n+1]/n from 1 to infinity, however the limit they compute is limn→∞ 1/n and NOT limn→∞ [(-1)n+1]/n.So did they just forget the absolute value sign in the definition (where they wrote the limit)? Or did they just feel showing the absolute value was trivial? Or do I just not understand the definition given? Does an always mean you just disregard the (-1)n?

Thanks for insight.
 
Physics news on Phys.org
\lim_{n \to \infty} a_n = 0 if and only if \lim_{n \to \infty} |a_n| = 0 so in that sense it doesn't matter.
 
  • Like
Likes   Reactions: Battlemage!
pasmith said:
\lim_{n \to \infty} a_n = 0 if and only if \lim_{n \to \infty} |a_n| = 0 so in that sense it doesn't matter.
Thought so. While testing power series end points I've been including the (-1)n terms in doing the limit test to see if they diverge, but this will save me some time. For example, I had one series which was (-1)nn3x3, and when testing the boundary point x = -1, I could have just looked at n3 instead of worrying about -1. Granted, it wasn't really much of a time saver since it's obvious that (-1)n*(-1)n is always positive, but I can see where it might help.

I appreciate the info.
 
I do have one other somewhat related question. For the integral test, can I also do the same with the (-1)n term? That would make things easier. For example, say I wanted to use the integral test to see if (-1)n/n2 converged. Could I just integrate 1/n2?

I mean since an infinite integral is really a limit, right?EDIT- I'm thinking NO, since 1/n diverges but (-1)n+1/n converges.
 
Battlemage! said:
I do have one other somewhat related question. For the integral test, can I also do the same with the (-1)n term? That would make things easier. For example, say I wanted to use the integral test to see if (-1)n/n2 converged. Could I just integrate 1/n2?
EDIT- I'm thinking NO, since 1/n diverges but (-1)n+1/n converges.
Besides that, you would have to explain what ##\int_0^{\infty} \frac{(-1)^{x+1}}{x}dx## actually means.
I mean since an infinite integral is really a limit, right?
Yes, ##\int_0^{\infty} f(x) dx = \lim_{a \rightarrow \infty} \int_0^{a} f(x) dx##
 

Similar threads

  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 11 ·
Replies
11
Views
6K
  • · Replies 37 ·
2
Replies
37
Views
5K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 16 ·
Replies
16
Views
4K
Replies
5
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 24 ·
Replies
24
Views
7K
  • · Replies 9 ·
Replies
9
Views
2K