I suppose the natural way to prove this is by induction. When I follow the rather natural steps $$\frac{(n+1)!}{(n+1)^{n+1}} = \frac{n!}{(n+1)^{n}} \leq \frac{n!}{(n)^{n}}$$ in order to apply the induction hypothesis, however, I obtain for $n+1$ exactly the same bound I assumed for $n$. This is not a problem if $n+1$ is odd, because then the floor of $n/2$ and $(n+1)/2$ coincide, but how can I "detect" the change in the case that $n + 1$ is even?
2026-03-27 11:48:55.1774612135
On
Prove $(n!)/n^n \leq 1/2^{k}$, where $k$ is the floor of $n/2$.
100 Views Asked by Bumbble Comm https://math.techqa.club/user/bumbble-comm/detail At
4
There are 4 best solutions below
0
On
Use $(n+1)^n = n^n \left(1+\frac{1}{n}\right)^n$ and prove $\left(1+\frac{1}{n}\right)^n\geq 2$ using binomial theorem.
Rather than proceeding by induction, note that \begin{align*} \frac{n!}{n^n} \leq \frac{k!}{n^k} \frac{n(n-1) \cdots (k+1)}{n^{n-k}} \leq \frac{k!}{n^k} = \left(\frac{1}{n}\right) \cdots \left(\frac{k}{n}\right) \leq \frac{1}{2^k}, \end{align*} since $k\leq 2n$.