Sparse matrices: convergence of the characteristic polynomial seen from infinity

Abstract: We prove that the reverse characteristic polynomial

d e t (I_{n} - z A_{n})

of a random

n i m e s n

matrix

A_{n}

with iid

m a t h r m B e r n o u l l i (d / n)

entries converges in distribution towards the random infinite product

p r o d_{e l l = 1}^{i} n f t y (1 - z^{e} l l)^{Y_{e} l l}

where

Y_{e} l l

are independent

m a t h r m P o i s s o n (d^{e} l l / e l l)

random variables. We show that this random function is a Poisson analog of more classical Gaussian objects such as the Gaussian holomorphic chaos. As a byproduct, we obtain new simple proofs of previous results on the asymptotic behaviour of extremal eigenvalues of sparse ErdH{o}s-R'enyi digraphs: for every

d > 1

, the greatest eigenvalue of

A_{n}

is close to

d

and the second greatest is smaller than

s q r t d

, a Ramanujan-like property for irregular digraphs. For

d < 1

, the only non-zero eigenvalues of

A_{n}

converge to a Poisson multipoint process on the unit circle. Our results also extend to the semi-sparse regime where

d

is allowed to grow to

i n f t y

with

n

, slower than

n^{o (1)}

. We show that the reverse characteristic polynomial converges towards a more classical object written in terms of the exponential of a log-correlated real Gaussian field, as in the dense case studied in a recent paper cite{bordenave2020convergence}. In the semi-sparse regime, the empirical spectral distribution of

A_{n} / s q r t d_{n}

converges to the circle distribution; as a consequence of our results, the second eigenvalue sticks to the edge of the circle.

This page was built for publication: Sparse matrices: convergence of the characteristic polynomial seen from infinity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6369151)