Necessary and sufficient conditions for almost sure convergence of the largest eigenvalue of a Wigner matrix (Q1123479): Difference between revisions

Let \(W=(X_{ij})\) be an infinite symmetric matrix. Suppose that the entries on the diagonal are iid, the entries off the diagonal are iid and they are independent random variables. Let \(W_ n=(X_{ij}:\) \(1\leq i,j\leq n).\) The author proves that necessary and sufficient conditions for \(\lambda_{\max}(W_ n/\sqrt{n})\to a\) a.s. are: \[ 1.\quad E(X^+_{11})^ 2<\infty,\quad 2.\quad E X^ 4_{12}<\infty,\quad 3.\quad E X_{12}\leq 0,\quad 4.\quad a=2\sigma,\quad \sigma^ 2=X^ 2_{12}. \]