0
$\begingroup$

How do I calculate the following limit: $\displaystyle\lim_{k \rightarrow \infty } \sqrt[k]{k(k+1)}$

The only limit identity that I know which closely resembles this is $\displaystyle\lim_{k \rightarrow \infty } \sqrt[k]{k}=1$.

Edit: This question came in context of finding the radius of convergence of $\displaystyle\sum_{k=1}^\infty \dfrac{2^k z^{2k}}{k^2+k}$.

Attempt: Clearly, the ratio test here is inconclusive. So, I tried Cauchy-Hadamard Formula.

For the general form of a power series, this says that $R=\displaystyle \dfrac{1}{\limsup_{k \rightarrow \infty} \sqrt[k]{|a_k|}}$.

But now $\displaystyle \lim_{k \rightarrow \infty} \sqrt[k]{\dfrac{2^k}{k^2+k}}=2 \lim_{k \rightarrow \infty} \sqrt[k]{\dfrac{1}{k^2+k}}=2$ as per the calculations done by @gimusi, @Key Flex.

Doubt: The answer at the back of the book gives $R=\frac{1}{\sqrt{2}}$.

$\endgroup$
3
$\begingroup$

We have that

$$ \sqrt[k]{k(k+1)}= \sqrt[k]{k} \,\sqrt[k]{k+1} \to 1 \cdot 1=1$$

indeed

$$\sqrt[k]{k+1}=e^{\frac{\ln k}{k+1}}\to e^0=1$$

and more in general for any polynomial $p_n(k)$ we have

$$\sqrt[k]{p_n(k)} \to 1$$

by the same proof.

$\endgroup$
  • $\begingroup$ How does this help? What is $\displaystyle\lim_{k \rightarrow \infty} \sqrt[k]{k+1}$? Equivalently, what is $\displaystyle\lim_{k \rightarrow \infty} \sqrt[k]{1+\frac{1}{k}}$ $\endgroup$ – user330477 Nov 4 '18 at 22:59
  • 1
    $\begingroup$ How do you evaluate $\lim_{k \rightarrow \infty } \sqrt[k]{k}=1$? $\endgroup$ – gimusi Nov 4 '18 at 23:00
  • $\begingroup$ @user330477 We have $\sqrt[k]{k+1}=e^{\frac{\log k}{k+1}}$? Does the $+1$ change something? $\endgroup$ – gimusi Nov 4 '18 at 23:01
  • $\begingroup$ @user330477 The key fact is that when k is large the "+1" is completely negligeble. $\endgroup$ – gimusi Nov 4 '18 at 23:04
  • 2
    $\begingroup$ Note that $$(1+\frac{1}{k})^{\frac{1}{k}}$$ is in the form $1^0$ which is not an indeterminate form. $\endgroup$ – gimusi Nov 4 '18 at 23:11
2
$\begingroup$

$\newcommand{\bbx}[1]{\,\bbox[15px,border:1px groove navy]{\displaystyle{#1}}\,} \newcommand{\braces}[1]{\left\lbrace\,{#1}\,\right\rbrace} \newcommand{\bracks}[1]{\left\lbrack\,{#1}\,\right\rbrack} \newcommand{\dd}{\mathrm{d}} \newcommand{\ds}[1]{\displaystyle{#1}} \newcommand{\expo}[1]{\,\mathrm{e}^{#1}\,} \newcommand{\ic}{\mathrm{i}} \newcommand{\mc}[1]{\mathcal{#1}} \newcommand{\mrm}[1]{\mathrm{#1}} \newcommand{\pars}[1]{\left(\,{#1}\,\right)} \newcommand{\partiald}[3][]{\frac{\partial^{#1} #2}{\partial #3^{#1}}} \newcommand{\root}[2][]{\,\sqrt[#1]{\,{#2}\,}\,} \newcommand{\totald}[3][]{\frac{\mathrm{d}^{#1} #2}{\mathrm{d} #3^{#1}}} \newcommand{\verts}[1]{\left\vert\,{#1}\,\right\vert}$ \begin{align} &\bbox[10px,#ffd]{\lim_{k \to \infty}\root[{\large k}]{k\pars{k + 1}}} = \exp\pars{\lim_{k \to \infty} {\ln\pars{k\bracks{k + 1}} \over k}} \\[5mm] = &\ \exp\pars{\lim_{k \to \infty}{\ln\pars{\bracks{k + 1}\bracks{k + 2}} - \ln\pars{k\bracks{k + 1}} \over \bracks{k + 1} - k}} \label{1}\tag{1} \\[5mm] = &\ \exp\pars{\lim_{k \to \infty}\ln\pars{1 + {2 \over k}}} = \bbx{\large 1} \end{align}

In expression \eqref{1}, I used the Stolz-Ces$\mathrm{\grave{a}}$ro Theorem.

$\endgroup$
1
$\begingroup$

HINT

$\lim_{k\to\infty}\sqrt[k]{a_k}=\lim_{k\to\infty}\frac{a_{k+1}}{a_k}$ if the second limit exists.

$\endgroup$
1
$\begingroup$

$$\lim_{k\rightarrow\infty}\sqrt[k]{k(k+1)}=\lim_{k\rightarrow\infty}((k^2+k)^{\frac1k})=\lim_{k\rightarrow\infty}\left(k^2\left(1+\dfrac1k\right)\right)^{\frac1k}=\lim_{k\rightarrow\infty}\left(1+\dfrac1k\right)^{\frac 1k}\cdot k^{\frac2k}=1\cdot1=1$$

$\endgroup$
  • $\begingroup$ You forgot the $\frac{1}{k}$ over $1+\frac{1}{k}$. This limit is what is causing me a lot of trouble. $\endgroup$ – user330477 Nov 4 '18 at 23:05
  • $\begingroup$ @user330477 I factored out $k^2$ from $k^2+k$, my computation is correct $\endgroup$ – Key Flex Nov 4 '18 at 23:07
  • 1
    $\begingroup$ No, your computation is not correct. The $\frac{1}{k}$ power should also be over $1+\frac{1}{k}$? $\endgroup$ – user330477 Nov 4 '18 at 23:08
  • $\begingroup$ @user330477 see the double parentheses, $(())$. It means $1+\dfrac1k$ is also included $\endgroup$ – Key Flex Nov 4 '18 at 23:13
  • 1
    $\begingroup$ @KeyFlex You lost an exponent here $$\ldots=\lim_{k\rightarrow\infty}\left(1+\dfrac1k\right)^{\color{red}{\frac1k}}\cdot k^{\frac2k}=1\cdot1=1$$ $\endgroup$ – gimusi Nov 4 '18 at 23:17
1
$\begingroup$

$(k)^{1/k}(k+1)^{1/(k+1)} \lt $

$(k(k+1))^{1/k} \lt$

$ (k)^{1/k}(2)^{1/k}(k^{1/k}).$

Take the limit.

$\endgroup$

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.