2
$\begingroup$

I am not sure whether the following inequality is true? Some small $n$ indicates it is true.

Let $n$ be a positive integer and $c\gt0$, then $$c^n(n!+c^n)\lt(n+c^2)^n.$$

$\endgroup$

3 Answers 3

5
$\begingroup$

Consider $n = 1$ and $c = 10$. Then, $c^n(n!+c^n) = 110$ while $(n+c^2)^n = 101$. You will be able to show a number of other similar counterexamples when $n=1$.

$\endgroup$
2
  • $\begingroup$ +1: Some small $n$ indicates it is true is not really true then :-) $\endgroup$
    – Aryabhata
    May 14, 2011 at 21:21
  • $\begingroup$ Sorry guys, I made a wrong formulation. I will update it soon. Can anyone help delete this post? $\endgroup$
    – Sunni
    May 14, 2011 at 21:32
3
$\begingroup$

I believe this is false.

Take $n=25, c = 5$.

I came up with this example by setting $c = \sqrt{n}$ and using Stirling's approximation formula for $n!$.

Wolfram alpha link showing the computation.

In fact for $c = \sqrt{n}$, we have that

$$\frac{c^n(n!+c^n)}{(n+c^2)^n} \sim \sqrt{2 \pi n}\ \left(\frac{n}{\sqrt{2e}}\right)^{n/2}$$

which goes to $\infty$ as $n \to \infty$.

$\endgroup$
2
  • $\begingroup$ Wow, you don't like trivial cases as $n=1$, do you. XD $\endgroup$ May 15, 2011 at 7:25
  • $\begingroup$ @Patrick: I actually ignored those, as OP said those seem to work... $\endgroup$
    – Aryabhata
    May 15, 2011 at 14:28
1
$\begingroup$

For $n=1$, you have $c+c^2 < 1+c^2$, which is equivalent to $c < 1$, hence always false because you supposed $c > 0$.

For $n=2$, you have $2c^2 + c^4 < c^4 +4c^2 + 4$, which is equivalent to $2c^2 > -4$ (always true).

Aryabhatta's example ensures us that we cannot conjecture that the inequality will always be true for $n > 2$.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .