Global eigenvalue distribution of matrices defined by the skew-shift (Q2236632)

One of the main purposes in random matrix theory is to prove the universality character of random models of matrices. In this article, the authors consider a Hermitian matrix of the form \[ \left( \begin{array}{cc} 0 & X \\ X^{*} & 0 \end{array} \right) \] where the entries of \(X\) are generated by sampling along the orbits of an ergodic dynamical system. The authors prove that for these type of matrices one still obtains the universality of the eigenvalues typical of random matrix models. This means that the distribution of the eigenvalues follow asymptotically the Wigner semicircle or the Marchenko-Pastur law. The analysis relies on the study of the moments of the distribution functions and their representation by means of graph theory.