Summing up all digamma roots

Calculus Level 5

x Ψ 1 x ( x 2 ) = 1 γ A B π C D 1 γ \Large \sum_{x \in \Psi} \frac{1}{x(x-2)}=\dfrac{1-\frac{\gamma^A}{B}-\frac{\pi^C}{D}}{1-\gamma}

Let Ψ \Psi is the set consisting of all roots of the digamma function .

Given that A , B , C A,B,C and D D are positive integers satisfying the equation above, find A + B + C + D A+B+C+D .

Notation : γ \gamma denotes the Euler-Mascheroni constant, γ = lim n ( ln ( n ) + k = 1 n 1 k ) 0.5772 \displaystyle\gamma = \lim_{n\to\infty} \left( - \ln(n) + \sum_{k=1}^n \dfrac1k \right) \approx 0.5772 .


The answer is 18.

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.

1 solution

Otto Bretscher
Jan 4, 2016

What an intriguing problem! It required a bit of detective work as I'm not exactly an expert in this field.

There is a useful formula for the digamma function, ψ ( z ) ψ ( z ) ψ ( z ) = 2 γ z k = 0 1 a k 2 a k z \frac{\psi'(z)}{\psi(z)}-\psi(z)=2\gamma-z\sum_{k=0}^{\infty}\frac{1}{a_k^2-a_kz} where the a k a_k are the zeros of ψ ( z ) \psi(z) . Evaluating this at z = 2 z=2 and observing that ψ ( 2 ) = 1 γ \psi(2)=1-\gamma and ψ ( z ) = n = 0 1 ( z + n ) 2 \psi'(z)=\sum_{n=0}^{\infty}\frac{1}{(z+n)^2} so ψ ( 2 ) = π 2 6 1 \psi'(2)=\frac{\pi^2}{6}-1 , we find that the sum we seek comes out to be 1 γ 2 2 π 2 12 1 γ \frac{1-\frac{\gamma^2}{2}-\frac{\pi^2}{12}}{1-\gamma}

so that the answer is 18 \boxed{18}

A link to the arxiv.org paper by Mezo, which presents this result, can be found on the Wikipedia page for the digamma function! The Wikipedia entry makes it clear that the paper is about summing roots of the digamma. Wikipedia beats Mathworld this time!

Mark Hennings - 5 years, 5 months ago

Log in to reply

What an observation! Great!

Aditya Kumar - 5 years, 5 months ago

Awesome job!

Here is a relevant paper . It has a derivation of that formula as well as a few other cool ones.

I wonder what other similar formulas we can find for other special functions. I don't think I've seen any like this for the Zeta function, I hope they're out there somewhere.

Isaac Buckley - 5 years, 5 months ago

Log in to reply

Thanks for the link!

Finding a similar formula for the zeta function might be your pathway to proving the Riemann hypothesis ;)

Otto Bretscher - 5 years, 5 months ago

Thanks for teaching something new!

Aditya Kumar - 5 years, 5 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...