∫ 0 1 [ ln ( Γ ( x ) ) ] 2 d x = B γ A + C γ ln D π + F π E + H G ln I J π + L π M ζ ′ ′ ( K ) − π P N γ ζ ′ ( O ) − π S ln ( Q π ) ζ ′ ( R )
If the equation above holds true for positive integers A , B , C , … , Q , R , S , then find the minimum value of A + B + C + D + E + F + G + H + I + J + K + L + M + N + O + P + Q + R + S .
Notations :
Γ ( ⋅ ) denotes the Gamma function .
γ denotes the Euler-Mascheroni constant , γ ≈ 0 . 5 7 7 2 .
ζ ′ and ζ ′ ′ denote the first and second derivative of the Riemann zeta function respectively.
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.
Brilliant!!
Log in to reply
My method was by using IBP. Then we can use the summation expansions of polygamma function. It takes little manipulation to get to the answer. Btw, what was your method?
Log in to reply
I used IBP too,it took me a while :( but at least i got the answer in the end :)
i found this in a paper :)
Right, so I don't really understand any of this, but I just wanted to say. WHAT WHAT WHAT Not even getting into the millions of unintelligible random symbols, THERE ARE 19 VARIABLES! WHAT EVEN IS THIS PROBLEM
Problem Loading...
Note Loading...
Set Loading...
In 1847 Kummer gave a Fourier series regarding Log Gamma which states ,
ln Γ ( x ) − 2 1 ln ( 2 π ) = − 2 1 ln ∣ 2 sin ( π x ) ∣ + 2 1 ( 1 − 2 x ) ( γ + ln ( 2 π ) ) + π 1 k = 2 ∑ ∞ k ln k sin ( 2 π k x )
Considering the symmetry of sin ( 2 π k x ) & a substitution of variable x → ( 1 − x ) for x ∈ ( 0 , 1 ) we have ,
ln Γ ( 1 − x ) − 2 1 ln ( 2 π ) = − 2 1 ln ∣ 2 sin ( π x ) ∣ − 2 1 ( 1 − 2 x ) ( γ + ln ( 2 π ) ) − π 1 k = 2 ∑ ∞ k ln k sin ( 2 π k x )
The above two results may be used to compute L G 2 = ∫ 0 1 ( ln ∣ Γ ( x ) ∣ ) 2 d x
Adding the two Blue results we get,
ln ∣ Γ ( x ) Γ ( 1 − x ) ∣ = ln ( 2 π ) − ln ∣ 2 sin ( π x ) ∣
Squaring and integrating both sides we get ,
∫ 0 1 ( ln Γ ( x ) Γ ( 1 − x ) ) ) 2 d x = ln 2 ( 2 π ) − π 1 L s 3 ( π )
L s 3 ( π ) denotes the Log-Sine Integrals .
As a standard result we may use L s 3 ( π ) = − 1 2 π 3 which yields ,
∫ 0 1 ( ln Γ ( x ) Γ ( 1 − x ) ) ) 2 d x = ln 2 ( 2 π ) + 1 2 π 2
Again by subtracting the Blue results we get ,
ln Γ ( 1 − x ) Γ ( x ) = ( 1 − 2 x ) ( γ + ln ( 2 π ) ) + π 2 k = 2 ∑ ∞ k ln k sin ( 2 π k x )
Which on squaring followed by integration and obviously some basic reductions gives us,
∫ 0 1 ( ln Γ ( 1 − x ) Γ ( x ) ) 2 d x = 3 1 ( γ + ln ( 2 π ) ) 2 + π 2 4 ( γ + ln ( 2 π ) ) k = 2 ∑ ∞ k 2 ln k + π 2 2 k = 2 ∑ ∞ k 2 ln 2 k
Using ζ ′ ( 2 ) = − k = 2 ∑ ∞ k 2 ln k & ζ ′ ′ ( 2 ) = k = 2 ∑ ∞ k 2 ln 2 k
∫ 0 1 ( ln Γ ( 1 − x ) Γ ( x ) ) 2 d x = 3 1 ( γ + ln ( 2 π ) ) 2 − π 2 4 ( γ + ln ( 2 π ) ) ζ ′ ( 2 ) + π 2 2 ζ ′ ′ ( 2 )
Now simply adding the two Green results and using the fact ∫ 0 1 ln Γ ( x ) d x = ∫ 0 1 ln Γ ( 1 − x ) d x we may deduce that ,
∫ 0 1 ( ln Γ ( x ) ) 2 d x = 3 ln 2 ( 2 π ) + 4 8 π 2 + 1 2 γ 2 + 3 γ ln ( 2 π ) − π 2 ln ( 2 π ) ζ ′ ( 2 ) − π 2 γ ζ ′ ( 2 ) + 2 π 2 ζ ′ ′ ( 2 )
So the answer is 9 7