1
$\begingroup$

Let $x_n$ be a sequence that converges to $a$. ($a$ final limit)

prove that if $\lim_\limits{n\to \infty}\frac{x_{n+1}}{x_n}=L$, then $|L| \le 1$

any suggestions guys? I thought about proving by contradiction by suggesting that $L>1$ then by the convergence test it would mean that $x_n$ diverges to infinity. but the problem that the convergence test works only for positive sequences.

$\endgroup$
  • 3
    $\begingroup$ If $a\neq 0$ then $L=1$. So the only case you are left with is $a=0$. $\endgroup$ – Thomas Andrews Dec 12 '15 at 20:54
  • $\begingroup$ You need to prove that $|L|\le 1$. If $a\ne 0$, then we know that $L=1$. So, analyze the specific case $a=0$ and show that $|L|\le 1$ also. $\endgroup$ – Mark Viola Dec 12 '15 at 21:05
  • 1
    $\begingroup$ If you are not trying to prove it, why are you asking this question? What? $\endgroup$ – Thomas Andrews Dec 12 '15 at 21:07
  • $\begingroup$ sorry, Typo mistake, I meant this is what Im trying to prove, the special case $a=0$ .. but couldn't. does it mean that its equals zero because $x_{n+1}$ converges to zero faster? $\endgroup$ – F1sargyan Dec 12 '15 at 21:09
  • 1
    $\begingroup$ If you knew the case $a\neq 0$, please put that in the question. You've caused me and Dr.MV to waste time telling you something you already knew. Help people help you. @F1sargyan $\endgroup$ – Thomas Andrews Dec 12 '15 at 21:11
1
$\begingroup$

HINT:

Assume that $\lim_{n\to \infty}x_n=0$ and $\lim_{n\to \infty}\frac{x_{n+1}}{x_n}=L$.

Show that for any given $\epsilon>0$, there exists a number $N(\epsilon)$ such that for $n>N$,

$$\min\left(|L-\epsilon|,|L+\epsilon|\right)^k\,|x_n| < |x_{n+k}| < \max\left(|L-\epsilon|,|L+\epsilon|\right)^k\,|x_n|$$

Note that as $k\to \infty$, $|x_{n+k}|\to 0$. Then conclude that $|L|\le 1$.

$\endgroup$
  • 2
    $\begingroup$ Assuming $x_n$ is positive here, and that $L$ is non-negative. $\endgroup$ – Thomas Andrews Dec 12 '15 at 21:14
  • $\begingroup$ @ThomasAndrews Yes. Good point. +1 for the catch. $\endgroup$ – Mark Viola Dec 12 '15 at 21:15
  • $\begingroup$ Ok, I got what u did there, but what assures the expression $min\left(|L-\epsilon|,|L+\epsilon|\right)^k$ to be bounded for example ? @Dr.MV $\endgroup$ – F1sargyan Dec 12 '15 at 22:14
  • $\begingroup$ If $|L|>1$, then there is a contradiction. $\endgroup$ – Mark Viola Dec 13 '15 at 2:50
1
$\begingroup$

Along the OP idea: assume that $\lim_{n\to \infty} \frac{ |x_{n+1}|}{|x_n|} > 1$. Then there exists $\epsilon > 0$ and $n_0$ so that $\frac{ |x_{n+1}|}{|x_n|} > 1+ \epsilon$ for all $n\ge n_0$, and so $|x_n| > (1+\epsilon)^{n-n_0}$ for all $n \ge n_0$, which implies $|x_n| \to \infty$.

$\endgroup$

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.