Density of eigenvalues and its perturbation invariance in unitary ensembles of random matrices (Q606669)

The aim of this paper is to study the eigenvalue distribution of unitary invariant ensembles of random matrices, which can be described by weight functions from the families of orthogonal polynomials. In contrast to the paper of \textit{P. G. Nevai} and \textit{J. S. Dehesa} [SIAM J.~Math.~Anal.~10, 1184--1192 (1979; Zbl 0434.33010)], where the moments of the density of eigenvalues have been calculated by studying the zero distribution of the orthogonal polynomials, here the moments are calculated directly without other knowledge than the three-term recurrence formula. It is proved that scaling eigenvalues converge weakly in probability and almost surely to the Nevai-Ullmann measure. It is shown that the density of eigenvalues is invariant under the polynomial perturbation of the weight function. Finally, a new way for the derivation of the differential equation of the density of eigenvalues is given.