On eigenvalues of the Brownian sheet matrix (Q6144446)

Motivated by the close connection between random matrix theory and interacting particle systems, it is natural to develop theory of random matrices with entries being random fields. Taking this into account, the authors derive a system of stochastic partial differential equations satisfied by the eigenvalues of the symmetric matrix whose entries are the values of the Brownian sheet. The tightness of the empirical spectral measures is obtained, and it guarantees, together with the classical Wigner's semicircle law, existence and uniqueness of the high-dimensional limit of the empirical spectral measures. It is proved that the limit measure satisfies a McKean-Vlasov equation, and its Stieltjes transform satisfies a generalized Burger's equation. It is noticed that the equations are obtained using the properties of the limit semicircle distribution and may have other equivalent forms. Since the Itô's formula for Brownian sheet contains additional terms, comparing to the corresponding formula for Dyson Brownian motion, the same situation occurs with the systems of the prelimit and limit equations.