Can you construct a sequence of real nonzero numbers whose sum converges to 0?

  • Context: Undergrad 
  • Thread starter Thread starter Bipolarity
  • Start date Start date
  • Tags Tags
    Converging Sum
Click For Summary

Discussion Overview

The discussion centers on the question of whether it is possible to construct a sequence of real nonzero numbers whose sum converges to 0. Participants explore various mathematical approaches and examples related to series and convergence.

Discussion Character

  • Exploratory
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant questions the existence of a sequence of nonzero numbers that sums to 0, suggesting that while sequences summing to nonzero values can be constructed, the case for 0 is less clear.
  • Another participant proposes a sequence using alternating signs and decreasing magnitudes: 1, -1, 0.5, -5, 0.25, -0.25, 0.125, -0.125, etc.
  • A request is made for an explicit representation of the proposed sequence using sigma notation.
  • A different participant presents a series defined by \(\sum^{\infty}_{n=1} (-1)^{n+1} \left(\frac{1}{2}\right)^{ \frac{2n-3+(-1)^{n+1}}{4}}\) and expresses belief that it converges to 0, seeking verification from others.
  • There is a discussion about the relationship between a series converging to a nonzero value and the implications for expressions involving that series, with some participants debating the necessity of expressing certain terms as nonzero.
  • One participant suggests a series involving \((1-λ)λ^n-(1-μ)μ^n\) for \(n \geq 0\), where \(λ\) is algebraic and \(μ\) is not, as a potential example.
  • Another participant proposes using a sequence that approaches 0 and a series defined by \(\sum_{x=0}^{\infty} (-1)^x b_x\), where \(b_x\) is defined in terms of the original sequence.
  • Further examples are provided, including a series involving positive terms that tend to zero paired with alternating signs, suggesting that there are multiple ways to construct such sequences.

Areas of Agreement / Disagreement

Participants express differing views on the existence of a sequence of nonzero numbers that sums to 0, with various proposed examples and methods. The discussion remains unresolved, with multiple competing approaches presented.

Contextual Notes

Some arguments depend on the definitions of series and convergence, and there are unresolved mathematical steps regarding the proposed sequences and their convergence properties.

Bipolarity
Messages
773
Reaction score
2
Does there exist a sequence of real nonzero numbers whose sum converges to 0?
I would think there isn't, but I'm interested in people's opinions and arguments.

For any nonzero m, a series of nonzero numbers whose sum converges to m can easily be constructed using the formula: \sum ^{\infty}_{n=1}m(0.5)^{n}

But that is for nonzero m, what if you wanted to construct a series whose sum converged to 0?

BiP
 
Physics news on Phys.org
1+-1+.5+-5+.25+-.25+.125+-.125+...
 
Vargo said:
1+-1+.5+-5+.25+-.25+.125+-.125+...

Can you find an explicit representation for that seqence (i.e. with sigma notation) ?

BiP
 
Eureka! I believe I found it!

\sum^{\infty}_{n=1} (-1)^{n+1} (\frac{1}{2})^{ \frac{2n-3+(-1)^{n+1}}{4}}

I believe it converges to 0, but can anyone verify this?

BiP
 
If \sum ^{\infty}_{n=1}m(0.5)^{n} converges to m, then shouldn't m-\sum ^{\infty}_{n=1}m(0.5)^{n} converge to 0?
 
Edgardo said:
If \sum ^{\infty}_{n=1}m(0.5)^{n} converges to m, then shouldn't m-\sum ^{\infty}_{n=1}m(0.5)^{n} converge to 0?

Yes, but m-\sum ^{\infty}_{n=1}m(0.5)^{n} is not a series... unless you can express it as one with nonzero terms.

BiP
 
Who cares about expressing it as one with nonzero terms? A series is a series is a series.
This is simple
How about
\sum_{n=0}^\infty \frac{(\pi)^{2n+1}}{(2n+1)!} (-1)^n
 
lurflurf said:
Who cares about expressing it as one with nonzero terms? A series is a series is a series.

The problem requires it.

BiP
 
Bipolarity said:
Yes, but m-\sum ^{\infty}_{n=1}m(0.5)^{n} is not a series... unless you can express it as one with nonzero terms.
How about (1-λ)λn-(1-μ)μn for n >= 0, where λ is algebraic and μ is not?
 
  • #10
How about taking a sequence (a_x)_x which satisfies \displaystyle \lim_{x\to\infty}a_x = 0 and then using the series \displaystyle \sum_{x=0}^{\infty} (-1)^x b_x, where the sequence b_x is defined as b_{2x} = b_{2x+1} = a_x?
 
  • #11
So you dislike the pi example and the usual example
\sum_{k=0}^\infty a_k b_x
where a_k is a sequence of positive numbers tending to zero and B_k is any sequence of -1 and 1 such that the series tends to zero.
What about any number of obvious examples such as
\sum_{k=0}^\infty (2k-1)\left(\frac{1}{3}\right)^k
 
Last edited:

Similar threads

  • · Replies 24 ·
Replies
24
Views
7K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 37 ·
2
Replies
37
Views
5K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 17 ·
Replies
17
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K