Local single ring theorem on optimal scale (Q2421816): Difference between revisions

Consider a \(N\times N\) random matrix of the form \[ X\equiv X_N=U\Sigma V^*, \tag{1}\] where \(U\equiv U_N\) and \(V\equiv V_N\) are two independent sequences of random matrices, which are both Haar distributed on either the unitary group \(U(N)\) of degree \(N\) or on the orthogonal group \(O(N)\) of degree \(N\). Moreover, let \(\Sigma\equiv \Sigma_N\) be a sequence of \(N\times N\) deterministic nonnegative definite diagonal matrices. By the single ring theorem in [\textit{A. Guionnet, M. Krishnapur} and \textit{O. Zeitount}, Ann. of Math. (2), 174, No. 2, 1189--1217 (2011; Zbl 1239.15018)] the empirical eigenvalue distribution of the matrix (1) converges weakly, in the limit of large \(N\), to a deterministic measure which is supported on a single ring centered at the origin in the complex plane \(\mathbb{C}\). In the interior of the single ring the convergence of the empirical eigenvalue distribution on the optimal local scale of order \(N^{-\frac{1}{2}+\varepsilon}\) and the optimal convergence rate are established. The same results hold true when \(U\) and \(V\) are Haar distributed on \(O(N)\).