Some trouble with a proof on $n!/(\sqrt{n})^n \geq 1$ Originally the problem is to prove that $n! \geq n^{n/2}$.
I reduced this to: $n! \geq (\sqrt{n})^n$ so that:
Prove that $\frac{n!}{(\sqrt{n})^n} \geq 1$.
Each term in $n!$ is divided by the $\sqrt{n}$ and the multiplication should leave it $\geq 1$.
Some advice.
 A: Consider the product
$$(1\cdot2\cdot 3\cdots n)( 1\cdot 2\cdot 3 \cdots n).$$
Divide these numbers into pairs, as in the "Baby Gauss" way of finding $1+2+3+\cdots +n$. Work from both ends in. Our product is
$$[1\cdot n][2\cdot (n-1)][3\cdot (n-2)]\cdots [n\cdot 1].$$
We have $n$ pairs, each with sum $n+1$.  In general, 
$$xy=\frac{(x+y)^2}{4}-\frac{(x-y)^2}{4}.$$
Let $x+y$ be fixed at $n+1$, and let $x$ and $y$ be positive integers. Then $xy$ is minimized when $|x-y|$ is as large as possible, that is, when $|x-y|=n-1$. So the minimum product of two paired numbers is $n$. It follows that 
$$(n!)^2 \ge n^n.$$
A: HINT: Note that $k \times (n+1-k) \geq n$ for $k \in \{1,2,\ldots,n\}$.
A: $$
s^{-s}=\frac{1}{\Gamma (s)}\int_{0}^{\infty }t^{s-1}e^{-st}dt
$$ 
(from here)
therefore
$(n!)^2 n^{-n}=\Gamma(n+1)^2\frac1{\Gamma(n)}\int \limits_0^\infty t^{n-1}e^{-nt}=n\Gamma(n+1)\int \limits_0^\infty t^{n-1}e^{-nt}\ge 1$
A: Hint (also, don't reexpress the problem like you did):
Suppose $n > 1$ is odd, that is, $n = 2k+1$ for some $k \geq 1$.
Then $n! = 1 * 2 * 3 * ... * (2k+1)$. Forget about the first multiplicand:
$n! = [2 * 3 * ... * (k+1)] * [(k+2) * ... * (2k+1)].$
Each bracket contains how many multiplicands? What else can you spot?
Then, a similar argument needs to be carried out for $n$ even.
A: $$n! \geq n^{n/2}\Leftrightarrow(n!)^{2}\geq n^{n}$$
Notice that 
$$(n!)^{2}
=\prod_{k=1}^{n}k\prod_{k=1}^{n}(n+1-k)
=\prod_{k=1}^{n}k(n+1-k)$$
and for each $k$, $k(n+1-k)-n=(k-1)(n-k)\geq0$, so $k(n+1-k)\geq n$ for each $k$, then $$(n!)^{2}=\prod_{k=1}^{n}k(n+1-k)\geq\prod_{k=1}^{n}n=n^{n}$$
A: The case $n=1$ is trivial, so let $n>1$.$$\frac{1}{n}\prod _{i=1}^{n-1}\left(\frac{1}{i}-\frac{1}{i+1}\right)=n!^{-2}\\\implies-\log n+\sum_{i=1}^{n-1}\log\left(\frac{1}{i}-\frac{1}{i+1}\right)=-2\log n!.$$ Thus we need to prove $$\sum_{i=1}^{n-1}\log\left(\frac{1}{i}-\frac{1}{i+1}\right)\leq (n-1)\log\frac{1}{n},$$ that follows immediately from Jensen's inequality.
EDIT:
A more straightforward path is simply to apply the AM-GM ineqauality. We need to show $$\prod_{i=1}^{n-1}\left(\frac{1}{i}-\frac{1}{i+1}\right)\leq \left(\frac{1}{n}\right)^{n-1}.$$ Applying AM-GM to the LHS yields $$\prod_{i=1}^{n-1}\left(\frac{1}{i}-\frac{1}{i+1}\right)\leq\left(\frac{1}{n-1}\sum_{i=1}^{n-1}\left(\frac{1}{i}-\frac{1}{i+1}\right)\right)^{n-1}=\left(\frac{1}{n-1}\left(1-\frac{1}{n}\right)\right)^{n-1},$$ which is the desired inequality.
A: Given the fact that (after proof) $ab \leq a + b - 1$. If we write $n!^2 = (n \cdot  1)\cdot (n-1)\cdot 2\cdot (n-2)\cdot 3 \cdots2 (n-1)\cdot (1\cdot n)$. $n\cdot 1\geq n+1-1\leq n$,  $(n-1)\cdot 2 \geq n - 1 + 2 -1\leq n,...$  According to the theorem each term $\geq n$.  So that $n!^2 \geq n^n$.  It follows that $n! \geq n^(n/2)$
