5
$\begingroup$

The series $\sum\limits_{n=1}^{\infty}\frac{n!}{n^n}$ is clearly positive and decreasing, but how does one go about integrating $\int\frac{x!}{x^x} dx$ ?

$\endgroup$
6
  • $\begingroup$ How would you define x! ? Do you mean $\Gamma(x)$? $\endgroup$
    – Vim
    Sep 13 '15 at 11:02
  • $\begingroup$ x! = x(x-1)(x-2)...(3)(2)(1) $\endgroup$
    – bard
    Sep 13 '15 at 11:05
  • 2
    $\begingroup$ it is invalid when x is not an integer. $\endgroup$
    – Vim
    Sep 13 '15 at 11:06
  • $\begingroup$ But how would you integrate that ?. I mean x! $\endgroup$
    – user210387
    Sep 13 '15 at 11:07
  • $\begingroup$ 'The series...is decreasing' That's not true: the series is increasing because every term is positive. What you mean is the sequence $\frac{k!}{k^k}$ is decreasing $\endgroup$
    – Alex
    Sep 13 '15 at 12:49
1
$\begingroup$

A very rough way of showing convergence is use Stirling's expansion of the numerator. Since $k! \sim (\frac{k}{e})^k \sqrt{2 \pi k}$ the summand becomes $a_k \sim \frac{\sqrt{2 \pi k}}{e^k}$ which you can then compare to the integral, $\Gamma(\frac{3}{2})$.

This, I repeat, is a pretty rough and not very elegant way of solving the problem. There are better ones too.

$\endgroup$
1
$\begingroup$

$\lim\limits_{n \to \infty}\sqrt[n]{\dfrac{n!}{n^n}}‎\sim \lim\limits_{n \to \infty} \dfrac{\dfrac{n}{e}}{n}=\dfrac{1}{e}<1$

$\endgroup$
2
  • $\begingroup$ Why is the nth root of n! n/e? $\endgroup$
    – bard
    Sep 13 '15 at 13:28
  • $\begingroup$ @bard, see "Stirling's approximation" $\endgroup$
    – user35603
    Sep 13 '15 at 13:48

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.