Zero-count conjecture for Fox's H-function

Let A=(A1,\textellipsis,Ap)\mathbf A=(A_1,\textellipsis,A_p) and B=(B1,\textellipsis,Bq)\mathbf B=(B_1,\textellipsis,B_q) have positive entries, let a=(a1,\textellipsis,ap)\mathbf a=(a_1,\textellipsis,a_p) and b=(b1,\textellipsis,bq)\mathbf b=(b_1,\textellipsis,b_q) have nonnegative entries, and let μ\mu, ρ\rho, and Q(t)Q(t) be the parameters and function used in the Fox HH-function representation. Assume

μ0,ρ1,i=1pAi=j=1qBj.\mu\ge0,\qquad \rho\le1,\qquad \sum_{i=1}^{p}A_i=\sum_{j=1}^{q}B_j.

Zero-count conjecture. The number of zeros of

Hq,pp,0(x(B,b)(A,a))H_{q,p}^{p,0}\left(x\left|\begin{array}{l}(\mathbf B,\mathbf b)\\(\mathbf A,\mathbf a)\end{array}\right.\right)

on (0,ρ)(0,\rho) is at most the number of zeros of Q(t)Q(t) on (0,1)(0,1).

The paper presents this as a stronger assertion than the known implication from nonnegativity of QQ to nonnegativity of the HH-function. It is supported by numerical evidence, while its general validity remains open.

Sources & referencesView supporting material

Primary source

Dmitrii Karp and Elena Prilepkina, “Completely monotonic gamma ratio and infinitely divisible H-function of Fox”, arXiv:1501.05388 (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.