Optimitation and perturbation analysis in the stationary distribution of a Markov Chain

Abstract

We considerer a stationery distribution of a finite, irreducible, homogeneus Markov Chain. Our aim is to pertub the transition probalities matriz using approximations to fin regions of feasibility and oiptimality for a given basis when the chain is optimized using linear programming. We also explore the application of perturbations bonds and analyze the effects of these on the construction of optimal policies.