A connection between the semigroup of the Cauchy process killed upon exiting a domain D and a mixed boundary value problem for the Laplacian in one dimension higher known as the mixed Steklov problem, was established in [6]. From this, a variational characterization for the eigenvalues λn, n ≥ 1, of the Cauchy process in D was obtained. In this paper we obtain a variational characterization of the difference between λn and λ1. We study bounded convex domains which are symmetric with respect to one of the coordinate axis and obtain lower bound estimates for λ ∗ −λ1 where λ ∗ is the eigenvalue corresponding to the “first ” antisymmetric eigenfunction for D. The proof is based on a variational characterization of λ ∗ − λ1 and on a weighted Poincaré–type inequality. The Poincaré inequality is valid for all α symmetric stable processes, 0 < α ≤ 2, and any other process obtained from Brownian motion by subordination. We also prove upper bound estimates for the spectral gap λ2 − λ1 in bounded convex domains.

Spectral gap for stable process on convex planar double symmetric domains

This note is an announcement of the results. The full version of this paper, including full proofs, will be available soon. 1.

Abstract. We study the spectral properties of the transition semigroup of the killed onedimensional Cauchy process on the half-line (0, ∞) and the interval (−1, 1). This process is related to the square root of one-dimensional Laplacian A = − − d2 dx2 with a Dirichlet exterior condition (on a complement of a domain), and to a mixed Steklov problem in the half-plane. For the half-line, an explicit formula for generalized eigenfunctions ψλ of A is derived, and then used to construct spectral representation of A. Explicit formulas for the transition density of the killed Cauchy process in the half-line (or the heat kernel of A in (0, ∞)), and for the distribution of the first exit time from the half-line follow. The formula for ψλ is also used to construct approximations to eigenfunctions of A in the interval. For the eigenvalues λn of A in the interval the asymptotic formula λn = nπ π 1 − + O ( ) is 2 8 n derived, and all eigenvalues λn are proved to be simple. Finally, efficient numerical methods of estimation of eigenvalues λn are applied to obtain lower and upper numerical bounds for the first few eigenvalues up to 9th decimal point. 1.