Ergodicity in parametric nonstationary Markov chains: an application to simulated annealing methods

摘要: A nonstationary Markov chain is weakly ergodic if the dependence of state distribution on starting vanishes as time tends to infinity. strongly it and converges in distribution. In this paper we show that two ergodicity concepts are equivalent for finite chains under rather general (and widely verifiable) conditions. We discuss applications probabilistic analyses search methods combinatorial optimization problems (simulated annealing).