Best-constant Stolarsky-mean approximation conjecture for the elliptic integral

Let r=1r2r^{\prime}=\sqrt{1-r^{2}}. For p0p\neq 0, let Sp,q(a,b)S_{p,q}(a,b) denote the Stolarsky mean, and let p01.763135p_{0}\approx 1.763135 be the unique solution on (0,9/4](0,9/4] of

(9/2pp)1/(2p9/2)=2π,\left(\frac{9/2-p}{p}\right)^{1/(2p-9/2)}=\frac{2}{\pi},

that is, θp0=2/π\theta_{p_{0}}=2/\pi. The best-constant Stolarsky-mean conjecture. There exists r0(0,1)r_{0}\in(0,1) such that

H(r)=1(2/π)E(r)1S9/2p0,p0(1,r)H(r)=\frac{1-(2/\pi)E(r)}{1-S_{9/2-p_{0},p_{0}}(1,r^{\prime})}

is strictly increasing on (0,r0)(0,r_{0}) and strictly decreasing on (r0,1)(r_{0},1). Consequently,

2πE(r)<S9/2p0,p0(1,r)\frac{2}{\pi}E(r)<S_{9/2-p_{0},p_{0}}(1,r^{\prime})

for r(0,1)r\in(0,1), with the best constant p0p_{0}. The paper proposes this claim as a conjecture, and the supplied text gives no resolution evidence.

Sources & referencesView supporting material

Primary source

Zhen-Hang Yang, “Very accurate approximations for the elliptic integrals of the second kind in terms of Stolarsky means”, arXiv:1508.05513 (2015).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.