Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

$$\lim_{n\to \infty} \left(1+\frac{1}{n!}\right)^{2n} $$

I've tried: $$\lim_{n\to \infty} \left(1+\frac{1}{n!}\right)^{2n} = \lim_{n\to \infty} e^{\ln{\left(1+\frac{1}{n!}\right)^{2n}}} = \lim_{n\to \infty} e^{2n\ \ln{\left(1+\frac{1}{n!}\right)}}$$ But I don't know how to work with the factorial

share|cite|improve this question
4  
$1/n!$ goes to 0 so hard that it hurts. It must be 1. I wouldn't use that argument in an exam though. – MyUserIsThis Mar 2 '13 at 19:50
up vote 14 down vote accepted

$$\lim_{n\to \infty} \left(1+\frac{1}{n!}\right)^{2n}=\lim_{n\to \infty}\left( \left(1+\frac{1}{n!}\right)^{n!}\right)^{\frac2{(n-1)!}}=e^0=1$$


Alternatively, Let $$A=\left(1+\frac{1}{n!}\right)^{2n}$$

So, $$\log A=2n\log \left(1+\frac{1}{n!}\right)\text { using }\log(1+x)=x-\frac{x^2}2+\frac{x^3}3-\frac{x^4}4+\cdots$$

which comes from the Taylor series which holds for $|x|<1$ as $n\to\infty, \frac1{n!}\to0\implies |\lim_{n\to\infty}\frac1{n!}|<1 $

$$=2n\left(\frac1{n!}-\frac{\frac1{(n!)^2}}2+\frac{\frac1{(n!)^3}}3+\cdots\right)$$

$$=\frac2{(n-1)!}-\frac1{n!(n-1)!}+\frac{2}{3(n-1)!(n!)^2}+\cdots$$

So, $$\lim_{n\to \infty}\log A=0\implies \lim_{n\to \infty}A=e^0=1 $$

share|cite|improve this answer
    
the exponent is $2n$ right? why did you write $\frac{2}{n}$ when taking the logarithm? – milo Mar 1 '13 at 14:34
    
@milo, rectifying – lab bhattacharjee Mar 1 '13 at 14:35
    
Thanks for both answers! – milo Mar 1 '13 at 14:39
    
@milo, thanks for your observation. – lab bhattacharjee Mar 1 '13 at 14:45
    
The first line would need some more explanation, like $\forall n (1+n^{-1})^n<e$ is also true - in some similar exercises these kind of limit-taking could lead to some nasty errors. – Ákos Somogyi Oct 26 '15 at 7:39

Since the exponential function is non-negative, we have for $x\geq 0$, $$0\leq \int_0^x\int_0^y \exp(z)\,dz\,dy=\exp(x)-x-1. $$ Plugging into the inequality $1+x \leq \exp(x)$ gives $$1\leq \left(1+\frac{1}{n!}\right)^{2n}\leq \exp\left(2n/n!\right)=\exp\left(2/(n-1)!\right)\to 1$$

share|cite|improve this answer
    
how do you prove the inequality? – milo Mar 1 '13 at 14:53
    
@milo I've added a proof of the inequality. – Byron Schmuland Mar 1 '13 at 16:52

Hint: $$\ln(1+x)=x+O(x^2)$$ as $x\to 0$. (So this limit is equal to 1.)

or you can use inequality $$\ln(1+x)\le x.$$

share|cite|improve this answer

Here, I complete @milo's try: (we use $\ln(1+x)\sim_0x) $$\lim_{n\to \infty} \left(1+\frac{1}{n!}\right)^{2n} = \lim_{n\to \infty} e^{\ln{\left(1+\frac{1}{n!}\right)^{2n}}} = \lim_{n\to \infty} e^{2n\ \ln{\left(1+\frac{1}{n!}\right)}}=\lim_{n\to \infty} e^{\frac{2n}{n!}}=e^0=1.$$

share|cite|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.