Gaussian fluctuations of eigenvalues in Wigner random matrices (Q967624): Difference between revisions

The paper deals with the fluctuations of eigenvalues from a class of Wigner random matrices that generalize the Gaussian orthogonal ensemble. The Gaussian orthogonal ensemble (GOE) is defined by the probability distribution on the space of \(n \times n\) real symmetric matrices \(H\) given by \[ P(dH)=C_n^{(\beta)}e^{-(\beta/2) Tr(H^2)}dH,\tag{1} \] where \(\beta=1\) and \(dH\) refers to the Lebesgue measure on the \(n(n+1)/2\) different elements of the matrix. On the other hand, the Gaussian unitary ensemble (GUE) is defined by the above probability distribution with \(\beta=2\), but on the space of \(n \times n\) Hermitian matrices. Finally, the Gaussian symplectic ensemble (GSE) is defined by the probability density given in (1) with \(\beta =4\), but on the space of \(n \times n\) quaternion self-dual Hermitian matrices. The author begins by considering an \(n \times n\) matrix from the GOE or the GSE and \(x_k\) denotes eigenvalue number \(k\). Under the condition that both \(k\) and \(n-k\) tend to infinity as \(n \rightarrow \infty\), he shows that \(x_k\) is normally distributed in the limit. The author also considers the joint limit distribution of eigenvalues \((x_{k_1}, \dots, x_{k_m})\) from the GOE or GSE, where \(k_1\), \(n-k_m\) and \(k_{i+1}-k_i\), \(1 \leq i \leq m-1\), tend to infinity with \(n\). The result in each case is an \(m\)-dimensional normal distribution. The main results of this paper extend \textit{J. Gustavsson}'s results [Ann. Inst. Henri Poincaré, Probab. Stat. 41, No. 2, 151--178 (2005; Zbl 1073.60020)] for the GUE to the GOE and GSE. Using recent universality results by \textit{T. Tao} and \textit{V. Vu} [ Commun. Math. Phys. 298, No. 2, 549--572 (2010; Zbl 1202.15038); \url{arXiv:0908.1982}], the author extends his results to a class of Wigner real symmetric matrices with non-Gaussian entries that have an exponentially decaying distribution and whose first four moments match the Gaussian moments.