Given a sequence $\{a_n\}_{n=1}^{\infty}$ and $c>0$ such that $a_{n+1} - a_n>c$ for every $n$, prove $\lim_{n\to\infty} a_n=\infty$.

I proved $a_n$ is monotonic increasing, but I'm having hard time proving it's unbounded.

Any ideas?


From the fact that $a_{n+1} > c+a_n$, you should prove (by induction) that $a_{n+1} > n\cdot c + a_1$.

Once you have that, proving that the sequence is unbounded should be a piece of pie, since for every $M\in\mathbb R$, you can set $n=\frac{M-a_1}{c}$ and get that $a_{n+1} > M.$

  • $\begingroup$ Yes, I've seen a similar solution, thought there was another way. Could you please explain how did you think of the inequality need to be proven ? $\endgroup$ – Itay4 Feb 14 '17 at 13:45
  • 1
    $\begingroup$ @Itay4 Well, after you just rearange $a_{n+1} - a_n>c$ into $a_{n+1} > c + a_n$, you can see that this equality actually means that "after each step, you move at least $c$ to the right". That's only one logical leap from thinking OK, so after $2$ steps, I am $2c$ to the right, and after $1000$ steps, I am $1000c$ to the right, so after $n$ steps, I will be $nc$ to the right". $\endgroup$ – 5xum Feb 14 '17 at 13:46
  • 1
    $\begingroup$ @Itay4 Then, you just put that into equation form and rigorously prove it. $\endgroup$ – 5xum Feb 14 '17 at 13:47
  • $\begingroup$ Got it, thanks ! $\endgroup$ – Itay4 Feb 14 '17 at 13:49

Prove $a_{n+1}>a_1+nc$ using induction or telescoping.


you can see that

$$ a_n-a_0= \sum_{i=1}^{n } a_i-a_{i-1}> c\sum_{i=1}^{n} 1 =cn $$

ie $$\infty = \lim nc+a_0 < \lim a_n $$ hence $a_n\to \infty$

  • $\begingroup$ Nice ! I know that weak inequality is preserved in limits, is it the same on strong one? $\endgroup$ – Itay4 Feb 14 '17 at 14:04
  • $\begingroup$ What do you mean by weak inequality in this context? Please clarify $\endgroup$ – Guy Fsone Feb 16 '17 at 11:41
  • $\begingroup$ I know that if $a\ \geq b$ then $lima\ \geq lim b$. Is it the same with "$>$" ? $\endgroup$ – Itay4 Feb 16 '17 at 14:49
  • $\begingroup$ $a<b\implies \lim a\le\lim b $. fpr instance $x>1\implies x< x^2$ but the limits coincide as $x\to 1.$ $\endgroup$ – Guy Fsone Feb 17 '17 at 11:34
  • $\begingroup$ Got it, thanks! $\endgroup$ – Itay4 Feb 17 '17 at 13:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.