1

What is the limit of the following sequence?. $$\frac{\left ( 2n \right )!}{(n!)^{2}4^{n}}$$ I guess the limit is zero but I don't know exactly how to prove it. Any help please.

Robert Z
  • 145,942
purecj
  • 344
  • 8
    Stirling's formula? – Angina Seng Mar 22 '18 at 07:11
  • @purecj Sorry, you are right. The ratio-test is inconclusive here. I am afraid that you need Stirling as suggested above or the estimate https://math.stackexchange.com/questions/58560/elementary-central-binomial-coefficient-estimates – Robert Z Mar 22 '18 at 07:50
  • The limit as $n$ goes to what? You should specify. Right now, the limit can by anything. – Enrico M. Mar 22 '18 at 08:16
  • @LordSharktheUnknown Stirling is so brutal! Ratio test si more elegant :) – user Mar 22 '18 at 08:40

3 Answers3

6

Because I think there are people interested in an elementary solution:

The ratio between terms $a_n=\binom{2n}{n}\frac{1}{4^n}$ is given by

$$\begin{array}{ll} \displaystyle \frac{a_{n+1}}{a_n} & =\frac{\displaystyle\frac{(2n+2)!}{(n+1)!(n+1)!}\frac{1}{4^{n+1}}}{\displaystyle\frac{(2n)!}{n!n!}\frac{1}{4^n}} \\[6pt] & \displaystyle = \frac{(2n+2)(2n+1)}{(n+1)(n+1)4} \\[3pt] & \displaystyle =\frac{2n+1}{2n+2} \\[2pt] & \displaystyle =1-\frac{1}{2n+2}. \end{array} $$

Therefore, the term $a_n$ is

$$\begin{array}{ll} a_n & \displaystyle =a_0\cdot\frac{a_1}{a_0}\cdot\frac{a_2}{a_1}\cdots\frac{a_n}{a_{n-1}} \\ & = \left(1-\frac{1}{2}\right)\left(1-\frac{1}{4}\right)\cdots\left(1-\frac{1}{2n}\right) \\ & \le \left(1-\frac{1}{3}\right)\left(1-\frac{1}{5}\right)\cdots\left(1-\frac{1}{2n+1}\right) \end{array} $$

and therefore

$$ \begin{array}{ll} a_n^2 & \le \left(1-\frac{1}{2}\right)\left(1-\frac{1}{3}\right)\cdots\left(1-\frac{1}{2n}\right)\left(1-\frac{1}{2n+1}\right) \\ & = \left(\frac{1}{2}\right)\left(\frac{2}{3}\right)\cdots\left(\frac{2n-1}{2n}\right)\left(\frac{2n}{2n+1}\right)=\frac{1}{2n+1}. \end{array} $$

From $a_n\le\frac{1}{\sqrt{2n+1}}$ we conclude the limit is $0$.

anon
  • 151,657
3

Simpler than this cannot be possible. By Stirling:

$$n! \approx \sqrt{2\pi n}n^n e^{-n}$$

Which of course has to be arranged to get an approximation for $(2n)!$ and for $(n!)^2$, which is trivial algebra. Then you get:

$$\frac{\sqrt{\pi } 2^{2 n+1} e^{-2 n} n^{2 n+\frac{1}{2}}}{2 \pi e^{-2 n} n^{2 n+1} 4^n}$$

Easy algebra will let you to rewrite this as a "goes like":

$$\color{red}{\sim \frac{1}{\sqrt{\pi } \sqrt{n}}}$$

Which as $n\to \text{whatever}\ $ you need, gives you the limit.

I think $n\to +\infty$ was understood, hence the limit is zero.

Enrico M.
  • 26,114
2

Alternative approach: since $\cos(x)=\frac{e^{ix}+e^{-ix}}{2}$ and $\int_{0}^{2\pi}e^{nix}e^{-mix}\,dx=2\pi\delta(m,n)$, as a consequence of the binomial theorem we have $$ a_n=\frac{1}{4^n}\binom{2n}{n}=\frac{1}{2\pi}\int_{0}^{2\pi}\cos(x)^{2n}\,dx. $$ By the dominated convergence theorem, the limit of the RHS as $n\to +\infty$ is zero.
So it is the limit of the LHS. Actually the integral representation also tells us that $\{a_n\}_{n\geq 1}$ is a log-convex sequence: $a_n a_{n+2} \geq a_{n+1}^2$ follows from Cauchy-Schwarz.


Here it comes the horrific overkill.
Since $\sum_{n\geq 0}a_n^3$ is a convergent series, equal to $\frac{1}{4\pi^3}\Gamma\left(\frac{1}{4}\right)^4$, we clearly have $\lim_{n\to +\infty}a_n=0$.

Jack D'Aurizio
  • 353,855