Circular law for noncentral random matrices

Abstract: Let

(X_{j k})_{j, k g e q 1}

be an infinite array of i.i.d. complex random variables, with mean 0 and variance 1. Let

l a_{n, 1}, ..., l a_{n, n}

be the eigenvalues of

(f r a c 1 s q r t n X_{j k})_{1 l e q j, k l e q n}

. The strong circular law theorem states that with probability one, the empirical spectral distribution

f r a c 1 n (d e_{l a_{n, 1}} + ... + d e_{l a_{n, n}})

converges weakly as

n o i n f t y

to the uniform law over the unit disc

z i n d C; | z | l e q 1

. In this short note, we provide an elementary argument that allows to add a deterministic matrix

M

to

(X_{j k})_{1 l e q j, k l e q n}

provided that

m a t h r m T r (M M^{*}) = O (n^{2})

and

m a t h r m r a n k (M) = O (n^{a} l)

with

a l < 1

. Conveniently, the argument is similar to the one used for the non-central version of Wigner's and Marchenko-Pastur theorems.

Cites Work

Cited In (11)

Uses Software

This page was built for publication: Circular law for noncentral random matrices