1
$\begingroup$

Recently I came across the Hardy-Ramanujan theorem, which states that the number of partitions of an integer $n$ denoted by $p(n)$ behaves asymptotically like $$e^{\pi\sqrt{2n/3}}.$$ This implies that $$\limsup p(n)^{1/n}=1.$$ I was wondering if it is relatively easier to derive the last statement or one has to go through the proof of Hardy and Ramanujan.

$\endgroup$
3
$\begingroup$

$$f(z)=\sum_{n\geq 0}p(n) z^n = \prod_{k\geq 1}\frac{1}{1-x^k} $$ is a holomorphic function in the region $|z|<1$ and the statement $$ \limsup p(n)^{1/n}=1$$ is equivalent to the statement "the radius of convergence of $f(z)$ at the origin is one", which is trivial.

| cite | improve this answer | |
$\endgroup$

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.