Interesting Zeta

Calculus Level 5

k = 2 ζ ( k ) 1 k + 1 = A B C B ln ( D π ) γ B \large \displaystyle \sum_{k=2}^{\infty} \frac{\zeta(k)-1}{k+1} = \frac{A}{B} - \frac{C}{B}\ln(D \pi) - \frac{\gamma}{B}

The above equation holds true for positive integers A A , B B , C C , and D D . Find A + B + C + D A+B+C+D .

Notation : γ 0.5772 \gamma \approx 0.5772 denotes the Euler-Mascheroni constant .


The answer is 8.

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

Rohan Shinde
Apr 25, 2019

First of all note the series ψ ( z + 1 ) + γ = k = 1 ( 1 ) k + 1 ζ ( k + 1 ) z k \displaystyle \psi(z+1)+\gamma=\sum_{k=1}^{\infty} (-1)^{k+1}\zeta(k+1) z^{k} For z < 1 \vert z\vert\lt 1 Now multiplying both sides by z z and then substituting z = x z=-x we get x ψ ( 1 x ) γ x = k 2 ζ ( k ) x k \displaystyle -x\psi(1-x)-\gamma x=\sum_{k\ge 2} \zeta(k)x^k Now we see that ξ = k 2 ζ ( k ) 1 k + 1 = k 2 ( ζ ( k ) 1 ) 0 1 x k d x \displaystyle \xi=\sum_{k\ge 2} \frac {\zeta(k)-1}{k+1}=\sum_{k\ge 2} (\zeta(k)-1)\int_0^1 x^k dx Interchanging summation and integration which can be justified by Tonelli's Theorem we get 0 1 ( ( k 2 ζ ( k ) x k ) k 2 x k ) d x \displaystyle \int_0^1 \left(\left(\sum_{k\ge 2} \zeta(k)x^k\right)-\sum_{k\ge 2} x^k\right)dx

Using the above obtained result alongwith the formula for infinite geometric progression we get ξ = 0 1 ( x ψ ( 1 x ) γ x x 2 1 x ) d x \xi=\int_0^1 \left(-x\psi(1-x)-\gamma x-\frac {x^2}{1-x}\right)dx ξ = 0 1 ( x ψ ( 1 x ) γ x ( x 1 + 1 1 x ) ) d x \xi=\int_0^1 \left(-x\psi(1-x)-\gamma x-\left(-x-1+\frac {1}{1-x}\right)\right)dx

Now properly splitting the integrals we have ξ = ξ 1 + ξ 2 \xi=\xi_1+\xi_2

Where ξ 1 = 0 1 ( x ψ ( 1 x ) + 1 1 x ) \displaystyle \xi_1=-\int_0^1 \left(x\psi(1-x)+\frac {1}{1-x}\right) And ξ 2 = 0 1 ( x + 1 γ x ) d x = 3 2 γ 2 \displaystyle \xi_2=\int_0^1 (x+1-\gamma x)dx=\color{#D61F06}{\frac 32-\frac {\gamma}{2}}

In ξ 1 \xi_1 we use the property that a b f ( x ) d x = a b f ( a + b x ) d x \displaystyle \int_a^b f(x)dx=\int_a^b f(a+b-x)dx to get ξ 1 = 0 1 ( ( 1 x ) ψ ( x ) + 1 x ) d x = 0 1 ( ψ ( x ) + 1 x ) d x + 0 1 x ψ ( x ) d x \displaystyle \xi_1=-\int_0^1\left((1-x)\psi(x)+\frac 1x\right)dx=-\int_0^1 \left(\psi(x)+\frac 1x\right)dx +\int_0^1 x\psi(x)dx

Now using the functional rule for digamma function that ψ ( z + 1 ) = ψ ( z ) + 1 z \psi(z+1)=\psi(z)+\frac 1z we get ξ 1 = 0 1 ψ ( x + 1 ) d x + 0 1 x ψ ( x ) d x = [ ln ( Γ ( x + 1 ) ) ] 0 1 + 0 1 x ψ ( x ) d x = 0 1 x ψ ( x ) d x \displaystyle \xi_1=-\int_0^1 \psi(x+1) dx +\int_0^1 x\psi(x)dx=\left[\ln(\Gamma(x+1))\right]_0^1+\int_0^1 x\psi(x) dx=\int_0^1 x\psi(x) dx

So now ξ 1 = 0 1 x ψ ( x ) d x \xi_1=\int_0^1 x\psi(x) dx Now using IBP once we get ξ 1 = 0 1 ln ( Γ ( x ) ) d x = 1 2 ln ( 2 π ) \displaystyle \xi_1=-\int_0^1 \ln(\Gamma(x)) dx=\color{#D61F06}{-\frac 12\ln(2\pi)} Where I use the standard result 0 1 ln ( Γ ( x ) ) d x = 1 2 ln ( 2 π ) \displaystyle \int_0^1 \ln(\Gamma(x)) dx=\frac 12\ln(2\pi) Which has been asked on Brilliant earlier also(I couldn't find the link to it).

So at the end we get ξ = ξ 1 + ξ 2 = 3 2 1 2 ln ( 2 π ) γ 2 \displaystyle \xi=\xi_1+\xi_2=\frac 32-\frac 12\ln(2\pi)-\frac {\gamma}{2}

Thus A + B + C + D = 8 A+B+C+D=\boxed {8}

Wow, same approach.....!! Btw, here is the link to the problem which you were looking for.....

Aaghaz Mahajan - 2 years, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...