0
$\begingroup$

I want to compute $$L=\lim_{n\to\infty}{\prod_{k=1}^{n}{\left(\frac{k}{n}\right)^{1/n}}}.$$ Let $P_n$ denote the $n$th partial product. Then, $$P_n=\left(\frac{n!}{n^n}\right)^{1/n}.$$ Now we can apply Stirling's approximation, $n!\sim\sqrt{2\pi n}\left(\frac{n}{e}\right)^n$ to obtain $$L=\frac{\sqrt{2\pi}}{e}\lim_{n\to\infty}{n^{\frac{1}{2n}}}=\frac{\sqrt{2\pi}}{e}.$$ However, the correct answer is apparently $L=e^{-1}$. Why am I off by a factor of $\sqrt{2\pi}$?

$\endgroup$

2 Answers 2

2
$\begingroup$

You would see the $2\pi$ appearing in the asymptotics but not in the limit $$P_n=\left(\frac{n!}{n^n}\right)^{1/n}=\frac{1}{e}+\frac{\log (2 \pi n)}{2 e n}+O\left(\frac{1}{n^2}\right)$$

$\endgroup$
1
$\begingroup$

Note that the $\sqrt{2\pi}$ in Stirling's approximation is just some constant, unlike the $e$ raised to the $n^{\text{th}}$ power. So, when you take $(2\pi)^{1/n},$ and let $n\rightarrow\infty,$ this $2\pi$ constant will go to 1 and not end up contributing to 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.