Fluctuations for matrix-valued Gaussian processes

Abstract: We consider a symmetric matrix-valued Gaussian process

Y^{(n)} = (Y^{(n)} (t); t g e 0)

and its empirical spectral measure process

m u^{(n)} = (m u_{t}^{(n)}; t g e 0)

. Under some mild conditions on the covariance function of

Y^{(n)}

, we find an explicit expression for the limit distribution of

Z_F^{(n)} := left( �ig(Z_{f_1}^{(n)}(t),ldots,Z_{f_r}^{(n)}(t)�ig) ; tge0 ight),

where

F = (f_{1}, d o t s, f_{r})

, for

r g e 1

, with each component belonging to a large class of test functions, and

Z_{f}^{(n)}(t) := nint_{mathbb{R}}f(x)mu_{t}^{(n)}( ext{d} x)-nmathbb{E}left[int_{mathbb{R}}f(x)mu_{t}^{(n)}( ext{d} x) ight].

More precisely, we establish the stable convergence of

Z_{F}^{(n)}

and determine its limiting distribution. An upper bound for the total variation distance of the law of

Z_{f}^{(n)} (t)

to its limiting distribution, for a test function

f

and

t g e q 0

fixed, is also given.

Recommendations

Cites Work

Cited In (7)

This page was built for publication: Fluctuations for matrix-valued Gaussian processes