English

The spectrum of heavy-tailed random matrices

Probability 2007-07-17 v1 Mathematical Physics math.MP

Abstract

Let XNX_N be an N\tsNN\ts N random symmetric matrix with independent equidistributed entries. If the law PP of the entries has a finite second moment, it was shown by Wigner \cite{wigner} that the empirical distribution of the eigenvalues of XNX_N, once renormalized by N\sqrt{N}, converges almost surely and in expectation to the so-called semicircular distribution as NN goes to infinity. In this paper we study the same question when PP is in the domain of attraction of an α\alpha-stable law. We prove that if we renormalize the eigenvalues by a constant aNa_N of order N1αN^{\frac{1}{\alpha}}, the corresponding spectral distribution converges in expectation towards a law μα\mu_\alpha which only depends on α\alpha. We characterize μα\mu_\alpha and study some of its properties; it is a heavy-tailed probability measure which is absolutely continuous with respect to Lebesgue measure except possibly on a compact set of capacity zero.

Keywords

Cite

@article{arxiv.0707.2159,
  title  = {The spectrum of heavy-tailed random matrices},
  author = {Gerard Ben Arous and Alice Guionnet},
  journal= {arXiv preprint arXiv:0707.2159},
  year   = {2007}
}
R2 v1 2026-06-21T08:58:22.208Z