Series $sum(-1)^{a(n)}frac1{a(n)}$ converging to infinity with $a:mathbb Ntomathbb N$ a bijection












1












$begingroup$


I'd love to get a hint or direction on this and figure out a solution myself:



Construct a bijection: $$: f:mathbb{N} rightarrow mathbb{N} $$
such that:
$$ sum_{n=0}^{infty} -1^{f(n)}frac{1}{f(n)} = infty $$
I thought of something like below to eliminate all series elements that are negative:
$$
f(n):=
left {
begin{array}{ll}
n^{odd} rightarrow n-1 \
n^{even} rightarrow n\
end{array}
right.
$$

But then $f(n)$ cannot be a bijection. Another way would be to transform the fraction $frac{1}{f(n)}$ into something that is divergent, but I don't see how that would be possible in an $mathbb{N} rightarrow mathbb{N}$ map.





EDIT: Deleted the suggestion that the resulting series would be a harmonic or geometric series.













share|cite|improve this question











$endgroup$












  • $begingroup$
    You mean $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty?$$
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 20:15










  • $begingroup$
    Yes, I did but I don't see what i did wrong.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:13






  • 1




    $begingroup$
    The series you are interested in are neither alternating nor harmonic.
    $endgroup$
    – Did
    Dec 10 '18 at 2:54
















1












$begingroup$


I'd love to get a hint or direction on this and figure out a solution myself:



Construct a bijection: $$: f:mathbb{N} rightarrow mathbb{N} $$
such that:
$$ sum_{n=0}^{infty} -1^{f(n)}frac{1}{f(n)} = infty $$
I thought of something like below to eliminate all series elements that are negative:
$$
f(n):=
left {
begin{array}{ll}
n^{odd} rightarrow n-1 \
n^{even} rightarrow n\
end{array}
right.
$$

But then $f(n)$ cannot be a bijection. Another way would be to transform the fraction $frac{1}{f(n)}$ into something that is divergent, but I don't see how that would be possible in an $mathbb{N} rightarrow mathbb{N}$ map.





EDIT: Deleted the suggestion that the resulting series would be a harmonic or geometric series.













share|cite|improve this question











$endgroup$












  • $begingroup$
    You mean $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty?$$
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 20:15










  • $begingroup$
    Yes, I did but I don't see what i did wrong.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:13






  • 1




    $begingroup$
    The series you are interested in are neither alternating nor harmonic.
    $endgroup$
    – Did
    Dec 10 '18 at 2:54














1












1








1


1



$begingroup$


I'd love to get a hint or direction on this and figure out a solution myself:



Construct a bijection: $$: f:mathbb{N} rightarrow mathbb{N} $$
such that:
$$ sum_{n=0}^{infty} -1^{f(n)}frac{1}{f(n)} = infty $$
I thought of something like below to eliminate all series elements that are negative:
$$
f(n):=
left {
begin{array}{ll}
n^{odd} rightarrow n-1 \
n^{even} rightarrow n\
end{array}
right.
$$

But then $f(n)$ cannot be a bijection. Another way would be to transform the fraction $frac{1}{f(n)}$ into something that is divergent, but I don't see how that would be possible in an $mathbb{N} rightarrow mathbb{N}$ map.





EDIT: Deleted the suggestion that the resulting series would be a harmonic or geometric series.













share|cite|improve this question











$endgroup$




I'd love to get a hint or direction on this and figure out a solution myself:



Construct a bijection: $$: f:mathbb{N} rightarrow mathbb{N} $$
such that:
$$ sum_{n=0}^{infty} -1^{f(n)}frac{1}{f(n)} = infty $$
I thought of something like below to eliminate all series elements that are negative:
$$
f(n):=
left {
begin{array}{ll}
n^{odd} rightarrow n-1 \
n^{even} rightarrow n\
end{array}
right.
$$

But then $f(n)$ cannot be a bijection. Another way would be to transform the fraction $frac{1}{f(n)}$ into something that is divergent, but I don't see how that would be possible in an $mathbb{N} rightarrow mathbb{N}$ map.





EDIT: Deleted the suggestion that the resulting series would be a harmonic or geometric series.










real-analysis sequences-and-series






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Dec 10 '18 at 12:36







lthz

















asked Dec 9 '18 at 20:01









lthzlthz

1246




1246












  • $begingroup$
    You mean $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty?$$
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 20:15










  • $begingroup$
    Yes, I did but I don't see what i did wrong.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:13






  • 1




    $begingroup$
    The series you are interested in are neither alternating nor harmonic.
    $endgroup$
    – Did
    Dec 10 '18 at 2:54


















  • $begingroup$
    You mean $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty?$$
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 20:15










  • $begingroup$
    Yes, I did but I don't see what i did wrong.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:13






  • 1




    $begingroup$
    The series you are interested in are neither alternating nor harmonic.
    $endgroup$
    – Did
    Dec 10 '18 at 2:54
















$begingroup$
You mean $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty?$$
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 20:15




$begingroup$
You mean $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty?$$
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 20:15












$begingroup$
Yes, I did but I don't see what i did wrong.
$endgroup$
– lthz
Dec 9 '18 at 22:13




$begingroup$
Yes, I did but I don't see what i did wrong.
$endgroup$
– lthz
Dec 9 '18 at 22:13




1




1




$begingroup$
The series you are interested in are neither alternating nor harmonic.
$endgroup$
– Did
Dec 10 '18 at 2:54




$begingroup$
The series you are interested in are neither alternating nor harmonic.
$endgroup$
– Did
Dec 10 '18 at 2:54










3 Answers
3






active

oldest

votes


















4












$begingroup$

The even positive integers can be partitioned into consecutive blocks $E_1 < E_2 < E_3 < cdots $ such that



$$sum_{nin E_k} 1/n > 2,,,text{for each } k.$$



Now think of listing $mathbb N$ as $E_1$ followed by $1,$ then $E_2$ followed by $3,$ then $E_3$ followed by $5,$ etc. This defines a bijection $f$ of $mathbb N$ onto $mathbb N$ for which



$$sum_{n=1}^{infty} (-1)^{f(n)}frac{1}{f(n)} ge 1+1+cdots = infty.$$






share|cite|improve this answer









$endgroup$













  • $begingroup$
    Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
    $endgroup$
    – lthz
    Dec 10 '18 at 12:34










  • $begingroup$
    Are you familiar with the Riemann rearrangement theorem?
    $endgroup$
    – zhw.
    Dec 10 '18 at 17:23










  • $begingroup$
    No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
    $endgroup$
    – lthz
    Dec 10 '18 at 17:50










  • $begingroup$
    Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
    $endgroup$
    – zhw.
    Dec 10 '18 at 18:01








  • 1




    $begingroup$
    I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
    $endgroup$
    – zhw.
    Dec 10 '18 at 20:54



















1












$begingroup$

I don't think you mean a geometric series because that's not what you've written so my answer will be pointing you towards a rearrangement of the series $$sum_{ngeq 1} frac{(-1)^n}{n} $$ which will diverge. We can do this because that series above is conditionally convergent.



Let $a_i = frac{1}{2i}$ and $b_j = -frac{1}{2j-1}$ for $i,jgeq 1$, then we will arrange the $a_i$ and $b_j$ so that their sum converge. Let $n_1$ be minimal such that $$a_1 + a_2 + cdots + a_{n_1} geq -b_1 + 1$$ Now let $n_2$ be minimal such that $$a_{n_1+1} + cdots + a_{n_2}geq -b_2 + 1$$



and continue in this way. Then $a_{n_{k-1}+1} + cdots+ a_{n_k} + b_k geq 1$ for all $k$ and so $$a_1 + a_2 + cdots + a_{n_1} + b_1 + a_{n_1 + 1} + cdots + a_{n_2} + b_2 + cdots $$ diverges. Hopefully this should be enough for you to construct your bijection.






share|cite|improve this answer









$endgroup$













  • $begingroup$
    "I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:18










  • $begingroup$
    This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
    $endgroup$
    – ODF
    Dec 9 '18 at 21:36










  • $begingroup$
    So you're admitting that you're not answering the question?
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:39










  • $begingroup$
    That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
    $endgroup$
    – ODF
    Dec 9 '18 at 21:40










  • $begingroup$
    Yes, ODF is right, I was mistaken. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:19



















1












$begingroup$

My answer addresses to the original question.



It's impossible to find a bijection $f: Bbb{N}toBbb{N}$ such that the alternating sum of the reciprocals is an alternating divergent geometric series.



To make $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty$$ an alternating divergent geometric series, you need $dfrac{a_{n+1}}{a_n} = -M < -1$ for some constant $M > 1$, where $a_n = (-1)^{f(n)}dfrac{1}{f(n)}$. (You need constant ratio $dfrac{a_{n+1}}{a_n}$ for geometric series, and if $0 < M < 1$, the series is absolutely convergent. If $M = 1$, $f$ is not a bijection.) The common ratio becomes $$-M = frac{a_{n+1}}{a_n} = (-1)^{f(n+1) - f(n)}frac{f(n)}{f(n+1)}.$$
Since the codomain of $f$ is $Bbb{N}$, $dfrac{f(n)}{f(n+1)} = M > 1$. It's easy to see that $f(n) = dfrac{1}{M^n} in (0,1)$, contradicting the condition that the codomain of $f$ is $Bbb{N}$.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks for answering the original question, which was misleading. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:29










  • $begingroup$
    @lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 22:45












  • $begingroup$
    okay noted, thanks!
    $endgroup$
    – lthz
    Dec 9 '18 at 22:52











Your Answer





StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3032904%2fseries-sum-1an-frac1an-converging-to-infinity-with-a-mathbb-n-to%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown

























3 Answers
3






active

oldest

votes








3 Answers
3






active

oldest

votes









active

oldest

votes






active

oldest

votes









4












$begingroup$

The even positive integers can be partitioned into consecutive blocks $E_1 < E_2 < E_3 < cdots $ such that



$$sum_{nin E_k} 1/n > 2,,,text{for each } k.$$



Now think of listing $mathbb N$ as $E_1$ followed by $1,$ then $E_2$ followed by $3,$ then $E_3$ followed by $5,$ etc. This defines a bijection $f$ of $mathbb N$ onto $mathbb N$ for which



$$sum_{n=1}^{infty} (-1)^{f(n)}frac{1}{f(n)} ge 1+1+cdots = infty.$$






share|cite|improve this answer









$endgroup$













  • $begingroup$
    Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
    $endgroup$
    – lthz
    Dec 10 '18 at 12:34










  • $begingroup$
    Are you familiar with the Riemann rearrangement theorem?
    $endgroup$
    – zhw.
    Dec 10 '18 at 17:23










  • $begingroup$
    No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
    $endgroup$
    – lthz
    Dec 10 '18 at 17:50










  • $begingroup$
    Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
    $endgroup$
    – zhw.
    Dec 10 '18 at 18:01








  • 1




    $begingroup$
    I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
    $endgroup$
    – zhw.
    Dec 10 '18 at 20:54
















4












$begingroup$

The even positive integers can be partitioned into consecutive blocks $E_1 < E_2 < E_3 < cdots $ such that



$$sum_{nin E_k} 1/n > 2,,,text{for each } k.$$



Now think of listing $mathbb N$ as $E_1$ followed by $1,$ then $E_2$ followed by $3,$ then $E_3$ followed by $5,$ etc. This defines a bijection $f$ of $mathbb N$ onto $mathbb N$ for which



$$sum_{n=1}^{infty} (-1)^{f(n)}frac{1}{f(n)} ge 1+1+cdots = infty.$$






share|cite|improve this answer









$endgroup$













  • $begingroup$
    Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
    $endgroup$
    – lthz
    Dec 10 '18 at 12:34










  • $begingroup$
    Are you familiar with the Riemann rearrangement theorem?
    $endgroup$
    – zhw.
    Dec 10 '18 at 17:23










  • $begingroup$
    No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
    $endgroup$
    – lthz
    Dec 10 '18 at 17:50










  • $begingroup$
    Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
    $endgroup$
    – zhw.
    Dec 10 '18 at 18:01








  • 1




    $begingroup$
    I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
    $endgroup$
    – zhw.
    Dec 10 '18 at 20:54














4












4








4





$begingroup$

The even positive integers can be partitioned into consecutive blocks $E_1 < E_2 < E_3 < cdots $ such that



$$sum_{nin E_k} 1/n > 2,,,text{for each } k.$$



Now think of listing $mathbb N$ as $E_1$ followed by $1,$ then $E_2$ followed by $3,$ then $E_3$ followed by $5,$ etc. This defines a bijection $f$ of $mathbb N$ onto $mathbb N$ for which



$$sum_{n=1}^{infty} (-1)^{f(n)}frac{1}{f(n)} ge 1+1+cdots = infty.$$






share|cite|improve this answer









$endgroup$



The even positive integers can be partitioned into consecutive blocks $E_1 < E_2 < E_3 < cdots $ such that



$$sum_{nin E_k} 1/n > 2,,,text{for each } k.$$



Now think of listing $mathbb N$ as $E_1$ followed by $1,$ then $E_2$ followed by $3,$ then $E_3$ followed by $5,$ etc. This defines a bijection $f$ of $mathbb N$ onto $mathbb N$ for which



$$sum_{n=1}^{infty} (-1)^{f(n)}frac{1}{f(n)} ge 1+1+cdots = infty.$$







share|cite|improve this answer












share|cite|improve this answer



share|cite|improve this answer










answered Dec 10 '18 at 0:57









zhw.zhw.

72.5k43175




72.5k43175












  • $begingroup$
    Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
    $endgroup$
    – lthz
    Dec 10 '18 at 12:34










  • $begingroup$
    Are you familiar with the Riemann rearrangement theorem?
    $endgroup$
    – zhw.
    Dec 10 '18 at 17:23










  • $begingroup$
    No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
    $endgroup$
    – lthz
    Dec 10 '18 at 17:50










  • $begingroup$
    Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
    $endgroup$
    – zhw.
    Dec 10 '18 at 18:01








  • 1




    $begingroup$
    I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
    $endgroup$
    – zhw.
    Dec 10 '18 at 20:54


















  • $begingroup$
    Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
    $endgroup$
    – lthz
    Dec 10 '18 at 12:34










  • $begingroup$
    Are you familiar with the Riemann rearrangement theorem?
    $endgroup$
    – zhw.
    Dec 10 '18 at 17:23










  • $begingroup$
    No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
    $endgroup$
    – lthz
    Dec 10 '18 at 17:50










  • $begingroup$
    Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
    $endgroup$
    – zhw.
    Dec 10 '18 at 18:01








  • 1




    $begingroup$
    I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
    $endgroup$
    – zhw.
    Dec 10 '18 at 20:54
















$begingroup$
Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
$endgroup$
– lthz
Dec 10 '18 at 12:34




$begingroup$
Yes, I see. It took me a while but I get it now. I would have never come up with that on my own. Can you give me a Theorem or something that made you think it through this way?
$endgroup$
– lthz
Dec 10 '18 at 12:34












$begingroup$
Are you familiar with the Riemann rearrangement theorem?
$endgroup$
– zhw.
Dec 10 '18 at 17:23




$begingroup$
Are you familiar with the Riemann rearrangement theorem?
$endgroup$
– zhw.
Dec 10 '18 at 17:23












$begingroup$
No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
$endgroup$
– lthz
Dec 10 '18 at 17:50




$begingroup$
No I was not actually and It would have been helpful. For some reason our prof decided to teach us the opposite, namely that any permutations of absolutely convergent series converge to the original limit. I actually have no idea why, because Riemanns Theorem seems far more important.
$endgroup$
– lthz
Dec 10 '18 at 17:50












$begingroup$
Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
$endgroup$
– zhw.
Dec 10 '18 at 18:01






$begingroup$
Actually the result on absolutely convergent series is probably more important; that is used all the time. I didn't use the RRT above, but once you see its proof, you enter a new world where groupings such as the one I used seem natural.
$endgroup$
– zhw.
Dec 10 '18 at 18:01






1




1




$begingroup$
I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
$endgroup$
– zhw.
Dec 10 '18 at 20:54




$begingroup$
I think we're fine. We have $$sum_{nin E_k} 1/n - frac{1}{2k-1}> 2-1,,,text{for each } k.$$
$endgroup$
– zhw.
Dec 10 '18 at 20:54











1












$begingroup$

I don't think you mean a geometric series because that's not what you've written so my answer will be pointing you towards a rearrangement of the series $$sum_{ngeq 1} frac{(-1)^n}{n} $$ which will diverge. We can do this because that series above is conditionally convergent.



Let $a_i = frac{1}{2i}$ and $b_j = -frac{1}{2j-1}$ for $i,jgeq 1$, then we will arrange the $a_i$ and $b_j$ so that their sum converge. Let $n_1$ be minimal such that $$a_1 + a_2 + cdots + a_{n_1} geq -b_1 + 1$$ Now let $n_2$ be minimal such that $$a_{n_1+1} + cdots + a_{n_2}geq -b_2 + 1$$



and continue in this way. Then $a_{n_{k-1}+1} + cdots+ a_{n_k} + b_k geq 1$ for all $k$ and so $$a_1 + a_2 + cdots + a_{n_1} + b_1 + a_{n_1 + 1} + cdots + a_{n_2} + b_2 + cdots $$ diverges. Hopefully this should be enough for you to construct your bijection.






share|cite|improve this answer









$endgroup$













  • $begingroup$
    "I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:18










  • $begingroup$
    This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
    $endgroup$
    – ODF
    Dec 9 '18 at 21:36










  • $begingroup$
    So you're admitting that you're not answering the question?
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:39










  • $begingroup$
    That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
    $endgroup$
    – ODF
    Dec 9 '18 at 21:40










  • $begingroup$
    Yes, ODF is right, I was mistaken. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:19
















1












$begingroup$

I don't think you mean a geometric series because that's not what you've written so my answer will be pointing you towards a rearrangement of the series $$sum_{ngeq 1} frac{(-1)^n}{n} $$ which will diverge. We can do this because that series above is conditionally convergent.



Let $a_i = frac{1}{2i}$ and $b_j = -frac{1}{2j-1}$ for $i,jgeq 1$, then we will arrange the $a_i$ and $b_j$ so that their sum converge. Let $n_1$ be minimal such that $$a_1 + a_2 + cdots + a_{n_1} geq -b_1 + 1$$ Now let $n_2$ be minimal such that $$a_{n_1+1} + cdots + a_{n_2}geq -b_2 + 1$$



and continue in this way. Then $a_{n_{k-1}+1} + cdots+ a_{n_k} + b_k geq 1$ for all $k$ and so $$a_1 + a_2 + cdots + a_{n_1} + b_1 + a_{n_1 + 1} + cdots + a_{n_2} + b_2 + cdots $$ diverges. Hopefully this should be enough for you to construct your bijection.






share|cite|improve this answer









$endgroup$













  • $begingroup$
    "I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:18










  • $begingroup$
    This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
    $endgroup$
    – ODF
    Dec 9 '18 at 21:36










  • $begingroup$
    So you're admitting that you're not answering the question?
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:39










  • $begingroup$
    That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
    $endgroup$
    – ODF
    Dec 9 '18 at 21:40










  • $begingroup$
    Yes, ODF is right, I was mistaken. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:19














1












1








1





$begingroup$

I don't think you mean a geometric series because that's not what you've written so my answer will be pointing you towards a rearrangement of the series $$sum_{ngeq 1} frac{(-1)^n}{n} $$ which will diverge. We can do this because that series above is conditionally convergent.



Let $a_i = frac{1}{2i}$ and $b_j = -frac{1}{2j-1}$ for $i,jgeq 1$, then we will arrange the $a_i$ and $b_j$ so that their sum converge. Let $n_1$ be minimal such that $$a_1 + a_2 + cdots + a_{n_1} geq -b_1 + 1$$ Now let $n_2$ be minimal such that $$a_{n_1+1} + cdots + a_{n_2}geq -b_2 + 1$$



and continue in this way. Then $a_{n_{k-1}+1} + cdots+ a_{n_k} + b_k geq 1$ for all $k$ and so $$a_1 + a_2 + cdots + a_{n_1} + b_1 + a_{n_1 + 1} + cdots + a_{n_2} + b_2 + cdots $$ diverges. Hopefully this should be enough for you to construct your bijection.






share|cite|improve this answer









$endgroup$



I don't think you mean a geometric series because that's not what you've written so my answer will be pointing you towards a rearrangement of the series $$sum_{ngeq 1} frac{(-1)^n}{n} $$ which will diverge. We can do this because that series above is conditionally convergent.



Let $a_i = frac{1}{2i}$ and $b_j = -frac{1}{2j-1}$ for $i,jgeq 1$, then we will arrange the $a_i$ and $b_j$ so that their sum converge. Let $n_1$ be minimal such that $$a_1 + a_2 + cdots + a_{n_1} geq -b_1 + 1$$ Now let $n_2$ be minimal such that $$a_{n_1+1} + cdots + a_{n_2}geq -b_2 + 1$$



and continue in this way. Then $a_{n_{k-1}+1} + cdots+ a_{n_k} + b_k geq 1$ for all $k$ and so $$a_1 + a_2 + cdots + a_{n_1} + b_1 + a_{n_1 + 1} + cdots + a_{n_2} + b_2 + cdots $$ diverges. Hopefully this should be enough for you to construct your bijection.







share|cite|improve this answer












share|cite|improve this answer



share|cite|improve this answer










answered Dec 9 '18 at 20:54









ODFODF

1,476510




1,476510












  • $begingroup$
    "I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:18










  • $begingroup$
    This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
    $endgroup$
    – ODF
    Dec 9 '18 at 21:36










  • $begingroup$
    So you're admitting that you're not answering the question?
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:39










  • $begingroup$
    That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
    $endgroup$
    – ODF
    Dec 9 '18 at 21:40










  • $begingroup$
    Yes, ODF is right, I was mistaken. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:19


















  • $begingroup$
    "I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:18










  • $begingroup$
    This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
    $endgroup$
    – ODF
    Dec 9 '18 at 21:36










  • $begingroup$
    So you're admitting that you're not answering the question?
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 21:39










  • $begingroup$
    That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
    $endgroup$
    – ODF
    Dec 9 '18 at 21:40










  • $begingroup$
    Yes, ODF is right, I was mistaken. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:19
















$begingroup$
"I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 21:18




$begingroup$
"I don't think you mean a geometric series because that's not what you've written" But OP has written the word "geometric series" both in the title and in the question body. Your series is not something that OP asks for.
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 21:18












$begingroup$
This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
$endgroup$
– ODF
Dec 9 '18 at 21:36




$begingroup$
This series is $sum (-1)^{f(n)} frac{1}{f(n)}$ for a bijection $f:mathbb{N}rightarrowmathbb{N}$ (which isn't a geometric series)
$endgroup$
– ODF
Dec 9 '18 at 21:36












$begingroup$
So you're admitting that you're not answering the question?
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 21:39




$begingroup$
So you're admitting that you're not answering the question?
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 21:39












$begingroup$
That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
$endgroup$
– ODF
Dec 9 '18 at 21:40




$begingroup$
That series is in the question? I think OP is mistaken as characterising it as a geometric sequence
$endgroup$
– ODF
Dec 9 '18 at 21:40












$begingroup$
Yes, ODF is right, I was mistaken. Sorry for that.
$endgroup$
– lthz
Dec 9 '18 at 22:19




$begingroup$
Yes, ODF is right, I was mistaken. Sorry for that.
$endgroup$
– lthz
Dec 9 '18 at 22:19











1












$begingroup$

My answer addresses to the original question.



It's impossible to find a bijection $f: Bbb{N}toBbb{N}$ such that the alternating sum of the reciprocals is an alternating divergent geometric series.



To make $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty$$ an alternating divergent geometric series, you need $dfrac{a_{n+1}}{a_n} = -M < -1$ for some constant $M > 1$, where $a_n = (-1)^{f(n)}dfrac{1}{f(n)}$. (You need constant ratio $dfrac{a_{n+1}}{a_n}$ for geometric series, and if $0 < M < 1$, the series is absolutely convergent. If $M = 1$, $f$ is not a bijection.) The common ratio becomes $$-M = frac{a_{n+1}}{a_n} = (-1)^{f(n+1) - f(n)}frac{f(n)}{f(n+1)}.$$
Since the codomain of $f$ is $Bbb{N}$, $dfrac{f(n)}{f(n+1)} = M > 1$. It's easy to see that $f(n) = dfrac{1}{M^n} in (0,1)$, contradicting the condition that the codomain of $f$ is $Bbb{N}$.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks for answering the original question, which was misleading. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:29










  • $begingroup$
    @lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 22:45












  • $begingroup$
    okay noted, thanks!
    $endgroup$
    – lthz
    Dec 9 '18 at 22:52
















1












$begingroup$

My answer addresses to the original question.



It's impossible to find a bijection $f: Bbb{N}toBbb{N}$ such that the alternating sum of the reciprocals is an alternating divergent geometric series.



To make $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty$$ an alternating divergent geometric series, you need $dfrac{a_{n+1}}{a_n} = -M < -1$ for some constant $M > 1$, where $a_n = (-1)^{f(n)}dfrac{1}{f(n)}$. (You need constant ratio $dfrac{a_{n+1}}{a_n}$ for geometric series, and if $0 < M < 1$, the series is absolutely convergent. If $M = 1$, $f$ is not a bijection.) The common ratio becomes $$-M = frac{a_{n+1}}{a_n} = (-1)^{f(n+1) - f(n)}frac{f(n)}{f(n+1)}.$$
Since the codomain of $f$ is $Bbb{N}$, $dfrac{f(n)}{f(n+1)} = M > 1$. It's easy to see that $f(n) = dfrac{1}{M^n} in (0,1)$, contradicting the condition that the codomain of $f$ is $Bbb{N}$.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thanks for answering the original question, which was misleading. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:29










  • $begingroup$
    @lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 22:45












  • $begingroup$
    okay noted, thanks!
    $endgroup$
    – lthz
    Dec 9 '18 at 22:52














1












1








1





$begingroup$

My answer addresses to the original question.



It's impossible to find a bijection $f: Bbb{N}toBbb{N}$ such that the alternating sum of the reciprocals is an alternating divergent geometric series.



To make $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty$$ an alternating divergent geometric series, you need $dfrac{a_{n+1}}{a_n} = -M < -1$ for some constant $M > 1$, where $a_n = (-1)^{f(n)}dfrac{1}{f(n)}$. (You need constant ratio $dfrac{a_{n+1}}{a_n}$ for geometric series, and if $0 < M < 1$, the series is absolutely convergent. If $M = 1$, $f$ is not a bijection.) The common ratio becomes $$-M = frac{a_{n+1}}{a_n} = (-1)^{f(n+1) - f(n)}frac{f(n)}{f(n+1)}.$$
Since the codomain of $f$ is $Bbb{N}$, $dfrac{f(n)}{f(n+1)} = M > 1$. It's easy to see that $f(n) = dfrac{1}{M^n} in (0,1)$, contradicting the condition that the codomain of $f$ is $Bbb{N}$.






share|cite|improve this answer











$endgroup$



My answer addresses to the original question.



It's impossible to find a bijection $f: Bbb{N}toBbb{N}$ such that the alternating sum of the reciprocals is an alternating divergent geometric series.



To make $$sum_{n=0}^{infty} (-1)^{f(n)}frac{1}{f(n)} = infty$$ an alternating divergent geometric series, you need $dfrac{a_{n+1}}{a_n} = -M < -1$ for some constant $M > 1$, where $a_n = (-1)^{f(n)}dfrac{1}{f(n)}$. (You need constant ratio $dfrac{a_{n+1}}{a_n}$ for geometric series, and if $0 < M < 1$, the series is absolutely convergent. If $M = 1$, $f$ is not a bijection.) The common ratio becomes $$-M = frac{a_{n+1}}{a_n} = (-1)^{f(n+1) - f(n)}frac{f(n)}{f(n+1)}.$$
Since the codomain of $f$ is $Bbb{N}$, $dfrac{f(n)}{f(n+1)} = M > 1$. It's easy to see that $f(n) = dfrac{1}{M^n} in (0,1)$, contradicting the condition that the codomain of $f$ is $Bbb{N}$.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Dec 9 '18 at 23:08

























answered Dec 9 '18 at 20:41









GNUSupporter 8964民主女神 地下教會GNUSupporter 8964民主女神 地下教會

12.8k72445




12.8k72445












  • $begingroup$
    Thanks for answering the original question, which was misleading. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:29










  • $begingroup$
    @lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 22:45












  • $begingroup$
    okay noted, thanks!
    $endgroup$
    – lthz
    Dec 9 '18 at 22:52


















  • $begingroup$
    Thanks for answering the original question, which was misleading. Sorry for that.
    $endgroup$
    – lthz
    Dec 9 '18 at 22:29










  • $begingroup$
    @lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
    $endgroup$
    – GNUSupporter 8964民主女神 地下教會
    Dec 9 '18 at 22:45












  • $begingroup$
    okay noted, thanks!
    $endgroup$
    – lthz
    Dec 9 '18 at 22:52
















$begingroup$
Thanks for answering the original question, which was misleading. Sorry for that.
$endgroup$
– lthz
Dec 9 '18 at 22:29




$begingroup$
Thanks for answering the original question, which was misleading. Sorry for that.
$endgroup$
– lthz
Dec 9 '18 at 22:29












$begingroup$
@lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 22:45






$begingroup$
@lthz It's noted. Please be aware that invalidating an existing answer through editing a question is not desirable. Users with over 2k reputation can edit (i.e. rollback) any existing posts. Therefore, such edit can be undone by any 2k user. However, to respond to questions "in the spirit in which they are asked", I'll leave your questoin as-is. However, please avoid that in the future.
$endgroup$
– GNUSupporter 8964民主女神 地下教會
Dec 9 '18 at 22:45














$begingroup$
okay noted, thanks!
$endgroup$
– lthz
Dec 9 '18 at 22:52




$begingroup$
okay noted, thanks!
$endgroup$
– lthz
Dec 9 '18 at 22:52


















draft saved

draft discarded




















































Thanks for contributing an answer to Mathematics Stack Exchange!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


Use MathJax to format equations. MathJax reference.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3032904%2fseries-sum-1an-frac1an-converging-to-infinity-with-a-mathbb-n-to%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

Quarter-circle Tiles

build a pushdown automaton that recognizes the reverse language of a given pushdown automaton?

Mont Emei