Descent Generating Polynomials and the Hermite–Biehler Theorem

Yang Arthur L. B.,Zhang Philip B.
DOI: https://doi.org/10.1007/s10801-021-01101-2
2022-01-01
Journal of Algebraic Combinatorics
Abstract:Brenti conjectured that, for any finite Coxeter group, the descent generating polynomial has only real zeros, and he left the type D case open. Dilks, Petersen, and Stembridge proposed a companion conjecture, which states that, for any irreducible finite Weyl group, the affine descent generating polynomial has only real zeros, and they left the type B and type D cases open. By developing the theory of $${\mathbf {s}}$$ -Eulerian polynomials, Savage and Visontai confirmed the type D case of the former conjecture and the type B case of the latter conjecture. In this paper, based on the Hermite–Biehler theorem and the theory of linear transformations preserving Hurwitz stability, we obtain the Hurwitz stability of certain polynomials related to the descent generating polynomials of type D, and thus give an alternative proof of Savage and Visontai’s results. This new approach also enables us to prove Hyatt’s conjectures on the interlacing property of half Eulerian polynomials of type B and type D, and to prove that the h-polynomial of certain subcomplexes of Coxeter complexes of type D has only real zeros. As one of the most important application of this approach, we completely confirm Dilks, Petersen, and Stembridge’s conjecture by proving the real-rootedness of the affine descent generating polynomials of type D.
What problem does this paper attempt to address?