An Error-Bound Theorem for Approximate Markov Chains
Nico M. Van DIJK · Probability in the Engineering and Informational Sciences · 1992
Recently an error-bound theorem was reported to conclude analytic error bounds for approximate Markov chains. The theorem required a uniform bound for marginal expectations of the approximate model. This note will relax this bound to steady state rather than marginal expectations as of practical interest (i) to simplify verification and/or (ii) to obtain a more accurate error bound. A communication model is studied in detail to support the results.