A stochastic ergodic theorem for superadditive processes

From MaRDI portal





Let V be the set of vectors \(v=(v_ 1,...,v_ d)\) with non negative integer coordinates and \(\{T_ v\), \(v\in V\}\) a semigroup of positive contractions in \(L_ 1\). A function \(F:V\to L_ 1\) is called superadditive process if \(F_ v\geq F_ u+T_ wF_{v-w}\) holds, whenever [0,v[ is the disjoint union of intervals [0,u[ and [w,v[. Put \(f_ v=(v_ 1v_ 2...v_ d)^{-1}F_ v.\) The authors pove: If \(\{T_ v\}\) is Markovian and \(\| f_ v\|_ 1\) bounded, and if F is bounded above by an additive process, then \((f_ v)\) converges stochastically for \(v\to \infty\). This extends results of the reviewer. (The proof can be reduced to these results, but the authors give an independent simple approach.)











This page was built for publication: A stochastic ergodic theorem for superadditive processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3333604)