For my 900 Followers and Comrades

Calculus Level 5

1 2016 2 2016 + 3 2016 4 2016 + 1^{2016}-2^{2016}+3^{2016}-4^{2016}+\cdots

The series above diverges, of course. Its Abel sum is defined as

A = lim x 1 n = 0 ( 1 ) n + 1 n 2016 x n A=\lim_{x\to 1^-}\sum_{n=0}^{\infty}(-1)^{n+1} n^{2016} x^n

Find A A to three significant figures. Enter 666 if you come to the conclusion that no such Abel sum A A exists.


Inspiration .


The answer is 0.00.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Mark Hennings
May 17, 2016

For any integer k 0 k \ge 0 we have n = 0 ( 1 ) n + 1 n k x n = ( x d d x ) k ( n = 0 ( 1 ) n + 1 x n ) = ( x d d x ) k 1 1 + x x < 1 \sum_{n=0}^\infty (-1)^{n+1}n^k x^n \; = \; \left(x\frac{d}{dx}\right)^k \left(\sum_{n=0}^\infty (-1)^{n+1}x^n\right) \; = \; -\left(x\frac{d}{dx}\right)^k \frac{1}{1+x} \qquad |x| < 1 Putting x = e y x = e^y for y < 0 y < 0 we see that n = 0 ( 1 ) n + 1 n k x n = ( d d y ) k 1 1 + e y \sum_{n=0}^\infty (-1)^{n+1}n^k x^n \; = \; -\left(\frac{d}{dy}\right)^k \frac{1}{1+e^y} and hence A k A_k , the Abel sum of n = 1 ( 1 ) n + 1 n k \sum_{n=1}^\infty (-1)^{n+1}n^k is equal to A k = lim x 1 n = 0 ( 1 ) n + 1 n k x n = lim y 0 ( d d y ) k 1 1 + e y A_k \; = \; \lim_{x\to1-}\sum_{n=0}^\infty (-1)^{n+1}n^k x^n \; = \; -\lim_{y\to0-}\left(\frac{d}{dy}\right)^k \frac{1}{1+e^y} and hence A k A_k is equal to k ! k! times the coefficient of y k y^k in the Maclaurin expansion of 1 1 + e y -\frac{1}{1+e^y} . Since 1 1 + e y = 1 2 m = 0 ( 1 ) m ( 2 2 m + 2 1 ) 2 2 m + 1 π 2 m + 2 ζ ( 2 m + 2 ) y 2 m + 1 \frac{1}{1+e^y} \; =\; \tfrac12 - \sum_{m=0}^\infty \frac{(-1)^m(2^{2m+2}-1)}{2^{2m+1}\pi^{2m+2}}\zeta(2m+2)y^{2m+1} we deduce that A k = 0 A_k = \boxed{0} for all even k k except k = 0 k=0 , and we have an expression for A k A_k for odd integers k k that can be written in terms of Bernoulli numbers.

Yes, that works!

Otto Bretscher - 5 years ago
Otto Bretscher
May 17, 2016

A n = L i n ( 1 ) = η ( n ) = ( 1 2 1 + n ) ζ ( n ) = ( 2 1 + n 1 ) B n + 1 n + 1 A_n=-Li_{-n}(-1)=-\eta(-n)=(1-2^{1+n})\zeta(-n)=\frac{(2^{1+n}-1)B_{n+1}}{n+1}

where L i Li is the polylogarithm, η ( s ) \eta(s) is Dirichlet's eta function, ζ ( s ) \zeta(s) is the zeta function, and B n B_n are the Bernoulli numbers. Thus A n = 0 A_n=\boxed{0} for positive even integers n n since the zeta function has its trivial zeroes at n -n .

See Comrade @Pi Han Goh 's inspirational problem for the first step.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...