Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Compute $$\lim_{n\to\infty} \left[\ln\left(\frac{1}{0!}+\frac{1}{1!}+\cdots+\frac{1}{n!}\right)\right]^n$$ If you have some nice proofs and you're willing to share them, then I thank you and you definitely have my upvote!

share|improve this question
4  
There is a routine proof, the missing tail of the sum is $\lt \frac{1}{nn!}$. –  André Nicolas Feb 10 '13 at 19:29
1  
more proofs... More than what? –  Did Feb 10 '13 at 19:30
1  
@did: I'll be glad even if MSE users offer me one proof. I'm usually interested in many proofs for each problem. The beauty of math is to watch/tackle a problem from many points of view. –  Chris's sis Feb 10 '13 at 19:34
    
@MaisamHedyelloo: yeah, right. Thanks! We get the case $1^{\infty}$. –  Chris's sis Feb 10 '13 at 19:37
    
Why one would want to gather proofs of THIS statement is a bit mysterious to me. The limit is 1, quite crude estimates yield it, and the result is not especially interesting nor meaningful nor esthetically pleasing. But maybe I am missing something? –  Did Feb 10 '13 at 19:40

2 Answers 2

up vote 5 down vote accepted

One could check that $$ \log\left(\sum\limits_{k=0}^n\frac{1}{k!}\right)=1+\alpha_n $$ where $$ \alpha_n=\log\left(1-e^{-1}\sum\limits_{k=n+1}^\infty\frac{1}{k!}\right) $$ Note that $$ 0\leq\lim\limits_{n\to\infty}n\alpha_n=\lim\limits_{n\to\infty}n(-e^{-1})\sum\limits_{k=n+1}^\infty\frac{1}{k!}\leq\lim\limits_{n\to\infty}n\frac{-1}{enn!}=0 $$ So $$ \lim\limits_{n\to\infty}\log^n\left(\sum\limits_{k=0}^n\frac{1}{k!}\right)= \lim\limits_{n\to\infty}\left(\left(1+\alpha_n\right)^{\frac{1}{\alpha_n}}\right)^{n\alpha_n}=e^0=1 $$

share|improve this answer
    
Thanks for your way! (+1) –  Chris's sis Feb 10 '13 at 20:11

The obvious way is to bound the missing tail of the series. That tail is less than a geometric series with sum $\frac{1}{nn!}$.

It follows that the thing $w$ inside the logarithm satisfies $$e-\frac{1}{nn!}\lt w\lt e.$$ Thus $$1+\log\left(1-\frac{1}{enn!}\right) \lt \log w \lt 1.$$ By the Taylor series for the logarithm, we have $$\log\left(1-\frac{1}{enn!}\right)=-\frac{1}{enn!}+o(1/nn!).$$ In particular, for large $n$, the logarithm is $\gt -\frac{2}{enn!}$. Now dealing with the $n$-th power is easy. The limit is $1$. There is an enormous amount of slack. A tail that is $O(1/n^2)$ would have been plenty good enough.

share|improve this answer
    
thanks for both hint and answer! (+1) –  Chris's sis Feb 10 '13 at 20:13

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.