Edge Universality for Inhomogeneous Random Matrices II: Markov Chain Comparison and Critical Statistics
This paper extends edge universality results for inhomogeneous random matrices to subcritical and critical sparsity regimes by establishing new Markov chain comparison conditions that link universal edge statistics to the comparability of underlying variance-profile Markov chains, thereby enabling the analysis of diverse models like random band and Hankel matrices beyond classical random matrix theory.