Error propogation formula of multi-level aggregation-disaggregation methods for non-symmetric problems

Abstract

Iterative aggregation-disaggregation methods for numerical computing of stationary probability distribution vectors of stochastic matrices are studied. The methods can use arbitrary numbers of levels and of smoothing steps. A formula for the error propagation is derived. Using this formula, some asymptotic convergence properties of these methods for non-symmetric problems are demonstrated.