The majorization conjecture for twisted root maps

About 23 years old · traced to

Let Cn\mathcal{C}_n be the class of polynomials of degree nn under consideration, let A′\mathcal{A}' be the class of operators defined in the paper, and write Z(P)Z(P) for the multiset of zeros of a polynomial PP. For a vector, ℜ\Re and ℑ\Im denote its componentwise real and imaginary parts, and ≺\prec denotes classical majorization.

Majorization conjecture. If T∈A′T\in\mathcal{A}' and n≥1n\ge 1, then for every P∈CnP\in\mathcal{C}_n at least one of the relations

ℜZ(P)≺ℜZ(T(P)),ℑZ(T(P))≺ℑZ(P)\Re Z(P)\prec \Re Z\bigl(T(P)\bigr),\qquad \Im Z\bigl(T(P)\bigr)\prec \Im Z(P)

is valid.

The conjecture proposes a global alternative between majorization of the real parts in the forward direction and majorization of the imaginary parts in the reverse direction for the operators under study. It is motivated by numerical experiments and by local results for real parameters, while the preceding counterexamples show that analogous complex statements involving convex-hull inclusion or stronger forms of majorization cannot hold generally.

References

Primary source

Julius Borcea, “Convexity properties of twisted root maps”, arXiv:math/0312321 (2007).

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.