A limit theorem for sets of stochastic matrices. (Q1430373)

The following fact about (row) stochastic matrices is an easy consequence of well known result: for each positive integer \(q=q(n)\) with the property that if \(A\) is any \(n\times n\) stochastic matrix then the sequence of matrices \(A^q, A^{2q}, A^{3q},\dots\) converges. The authors show a generalization of this for sets of stochastic matrices under the Hausdorff metric. Let \(d\) be any metric inducing the standard topology on \(M_n(\mathbb R)\), the set of \(n\times n\) real matrices. For \(A\in M_n(\mathbb R)\supset \mathcal B\), define \(d(A, \mathcal B):=\inf_{B\in\mathcal B}d(A, B)\). For \(\mathcal A, \mathcal B\subset M_n(\mathbb R)\), define \(d^+(\mathcal A, \mathcal B):=\sup_{A\in\mathcal A}d(A,\mathcal B)\), and \(d(\mathcal A, \mathcal B):= \max(d^+(\mathcal A, \mathcal B), d^+(\mathcal B, \mathcal A))\). The restriction to stochastic matrices is the Hausdorff metric on the set of subsets of stochastic matrices. If \(\mathcal A\) is a set of stochastic matrices, \(k\) a positive integer, then define \(\mathcal A^{(k)}\) to be the set of all matrices expressible as a product of a sequence of \(k\) matrices from \(\mathcal A\). The authors show that for each positive integer \(n\) there is a positive integer \(p(n)\) such that if \(\mathcal A\) is any subset of \(n\times n\) stochastic matrices then the sequence of subsets \(\mathcal A^{(p)},\mathcal A^{(2p)}, {\mathcal A}^{(3p)},\dots\) converges with respect to the Hausdorrf metric.