1. TWO-TIME-SCALE MARKOV CHAINS AND APPLICATIONS TO QUASI-BIRTH-DEATH QUEUES.
- Author
-
Yin, G. and Hanqin Zhang
- Subjects
- *
MARKOV processes , *STOCHASTIC processes , *BIRTH & death processes (Stochastic processes) , *COMPUTATIONAL complexity , *ASYMPTOTIC expansions , *STOCHASTIC convergence - Abstract
Aiming at reduction of complexity, this work is concerned with two-time-scale Markov chains and applications to quasi-birth-death queues. Asymptotic expansions of probability vectors are constructed and justified. Lumping all states of the Markov chain in each subspace into a single state, an aggregated process is shown to converge to a continuous-time Markov chain whose generator is an average with respect to the stationary measures. Then a suitably scaled sequence is shown to converge to a switching diffusion process. Extensions of the results are presented together with examples of quasi-birth-death queues. [ABSTRACT FROM AUTHOR]
- Published
- 2004
- Full Text
- View/download PDF