Numerical Solution of Large Finite Markov Chains by Algebraic Multigrid Techniques

Iterative aggregation/disaggregation procedures are a convenient numerical solution method for computing the stationary distribution vector of an ergodic homogeneous Markov chain with a finite state space. We show the equivalence of this method and a two-level multigrid method. Based on error results of the A/D-method, we provide an error analysis of an efficient multigrid variant of the multiplicative Schwars-iteration method. Furthermore, we apply these results to a multigrid version of the replacement process approach developed by Sumita and Rieders.