-2

Need help to find the limit of $$\lim _{ n\rightarrow \infty }{ \sum _{ k=1 }^{ n }{ \frac { \sqrt { { n }^{ 2 } - { k }^{ 2 } } } { { n }^{ 2 } } } }$$

Matthew Cassell
  • 4,248
  • 4
  • 21
  • 30
Ruslan
  • 19

2 Answers2

1

Hint:
$$\lim _{ n\rightarrow \infty }{ \sum _{ k=1 }^{ n }{ \frac { \sqrt { { n }^{ 2 } - { k }^{ 2 } } } { { n }^{ 2 } } } }=\lim_{n \to \infty } \sum _{ k=1 }^{ n }\frac{1}{n}\sqrt{1-\frac{k^2}{n^2}}$$

Replace $\frac{1}{n}$ with ${dx}$ and $\frac{k}{n}$ with $x$

$$ \lim _{ n\rightarrow \infty }\sum _{ k=1 }^{ n }\frac{1}{n}\sqrt{1-\frac{k^2}{n^2}}=\int_0^1\sqrt{1-x^2}{dx}$$

I obtained the bounds 0 and 1 using

$$\int_a^bf(x)\,\mathrm{d}x=\lim_{n\to\infty}\sum_k^n\frac{b-a}nf\left(k\frac{b-a}{n}\right)$$

  • I'd like to note that you can justify this with $\int_a^bf(x),\mathrm{d}x=\lim_{n\to\infty}\sum_k^n\frac{b-a}nf\left(k\frac{b-a}{n}\right)$ (hence where the bounds $0$ and $1$ come from) – Jam Sep 01 '18 at 10:09
  • Yes that's what I did. dodo you want me to include this in the answer? – Deepesh Meena Sep 01 '18 at 10:10
  • You could do - the reason I pointed it out is that the bounds seem to come out of nowhere – Jam Sep 01 '18 at 10:11
  • But one could also found the bounds like this $\frac{k}{n}$ is replaced by x but k can go max up to n thus x can go up to 1 also k is minimum 1 thus minimum x is 1/n but $n \to \infty$ thus minimum x is 0 – Deepesh Meena Sep 01 '18 at 10:15
1

As noticed by Riemann sum we obtain

$$\lim _{ n\rightarrow \infty }\sum _{ k=1 }^{ n }\frac{1}{n}\sqrt{1-\frac{k^2}{n^2}}=\int_0^1\sqrt{1-x^2}{dx}=\frac{\pi}4$$

As an alternative, just to explore other ways, since for a fixed $k$ by binomial series we have

$$\sqrt{1-\frac{k^2}{n^2}}=\sum_{j=0}^\infty \binom{\frac12}{j}\left(-\frac{k^{2}}{n^{2}}\right)^j $$

and therefore by Faulhaber's formula

$$\sum _{ k=1 }^{ n }\frac{1}{n}\sqrt{1-\frac{k^2}{n^2}} =\sum_{j=0}^\infty \left[\binom{\frac12}{j}\frac{(-1)^j}{n^{2j+1}}\sum_{k=1}^n k^{2j}\right] \to\sum_{j=0}^\infty \frac{(-1)^j\binom{\frac12}{j}}{{2j+1}}$$

and since for the binomial coefficient with $n=1/2$ we have

$${{\frac12}\choose{j}}={{2j}\choose{j}}\frac{(-1)^{j+1}}{2^{2j}(2j-1)}\implies \binom{\frac12}{j}\frac{(-1)^j}{{2j+1}}=\frac{\binom{2j}{j}}{2^{2j}(1-4j^2)}$$

we obtain that

$$\sum_{j=0}^\infty \frac{(-1)^j\binom{\frac12}{j}}{{2j+1}}=\sum_{j=0}^\infty \frac{\binom{2j}{j}}{2^{2j}(1-4j^2)}=\frac{\pi}4$$

user
  • 154,566
  • How do you justify that last equality to $\pi/4$? – Jam Sep 01 '18 at 10:39
  • @Jam I have not anyidea about that! I'm assuming it is true from the result obtained by Riemann sum. I just wanted explore this other ways. Numerical extimation confirm the result. Do you have any idea how to derive the result in a different way without using Riemann sum? – user Sep 01 '18 at 10:42
  • If you format your links as [link name](link address), you can fit the whole URL in the link - otherwise I can't click on it. – Jam Sep 01 '18 at 10:47
  • And I'm not sure about how to prove the equality to $\pi/4$. I'd got as far as you but haven't thought of how to do the last step. – Jam Sep 01 '18 at 10:47
  • @Jam I can't link the second sum, I don't know why it doesn't work. – user Sep 01 '18 at 10:52
  • No worries. I'm not sure I'd be able to help anyway haha. But for the record, you could try putting the link through bit.ly and then using that as the link for a comment. – Jam Sep 01 '18 at 10:53
  • @Jam Good idea! sum 2 – user Sep 01 '18 at 10:55
  • @Jam I might create an OP to ask a question about that! – user Sep 01 '18 at 11:07