2
$\begingroup$

Give an example of a convergent sequence $\{a_n\}$ and divergent sequence $\{b_n\}$ such that $\{a_n+b_n\}$ is a convergent series.

I've been trying to solve this question for a couple days now and have been struggling, if anyone could give me a hint or show me how you got your answer as I feel this isn't solvable but the question says that I must have an example. Thank you in advance, Math Student :)

$\endgroup$
4
  • 5
    $\begingroup$ Who asked you that? You're unlikely to find an example, it is not possible. $\endgroup$
    – Clement C.
    Commented Jan 23, 2017 at 21:32
  • 4
    $\begingroup$ As pointed out by Nayuki, your title and body are asking subtly but crucially different questions. Is the sum supposed to be a convergent sequence, or a convergent series? (And for that matter, what about $\{a_n\}$ and $\{b_n\}$?) $\endgroup$ Commented Jan 24, 2017 at 3:31
  • $\begingroup$ If only 1 + 2 + 3 + ... = -1/12... $\endgroup$
    – Andrew T.
    Commented Jan 24, 2017 at 6:34
  • $\begingroup$ You still need to decide whether you mean "sequence" or "series"; I merely improved your LaTeX and title. $\endgroup$
    – user21820
    Commented Jan 24, 2017 at 7:40

3 Answers 3

14
$\begingroup$

Assume ${a_n + b_n}$ converges. Since ${a_n}$ converges, ${a_n + b_n - a_n}$ converges, contradicting the fact that ${b_n}$ does not converge.

$\endgroup$
3
  • 5
    $\begingroup$ Does your answer take into account the fact that in the question, $a_n$ is a convergent sequence, $b_n$ is a divergent sequence, and $a_n+b_n$ is a convergent series? $\endgroup$
    – Nayuki
    Commented Jan 24, 2017 at 3:06
  • 2
    $\begingroup$ @Nayuki If by convergent series you mean that the series over $a_n+b_n$, i.e. $\sum_{i=1}^\infty (a_i+b_i)$, converges, then this implies that the sequence $(a_n+b_n)$ converges. $\endgroup$ Commented Jan 24, 2017 at 5:55
  • $\begingroup$ {$a_n$},{$b_n$}, and {$a_n$+$b_n$} are all sequences, sorry for leaving that out $\endgroup$ Commented Jan 24, 2017 at 22:52
4
$\begingroup$

This is not possible. Suppose it was, then we have a convergent sequence ${a_n+b_n}$, and ${a_n}$. We know that the difference of convergent sequences is itself, convergent. This means if ${a_n+b_n}$ converges, and ${a_n}$ converges, then ${a_n+b_n-a_n}$ converges, but this means that ${b_n}$ is convergent, which contradicts our hypothesis, so no, this cannot be done.

You can however have two divergent sequences sum to a convergent one. Just take ${a_n}=(n)$ and ${b_n}=(-n)$ which gives us ${a_n+b_n}=(0)$, and the zero sequence is a constant sequence, which is trivially convergent.

$\endgroup$
0
$\begingroup$

Since you were seeking an example and one could not be found, but neither answer has been accepted, I can offer this as an example of one... although it may not be appropriate in the context of the tag real analysis.

Let $\forall i:a_i,b_i\in\Bbb Z[\frac12]/\Bbb Z$ the dyadic rationals in the interval $(0,1]$

Then consider the sequences:

The convergent sequence $b_n=\frac12,\frac34,\frac78,\frac{15}{16}\ldots\to1$

and the divergent sequence $1-b_n$

Their sum is the constant sequence $1,1,1\ldots$ which converges in $(0,1]$.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .