
用耦合方法研究马氏链f-指数遍历
Study of f-exponent Ergodic of Markov Chains by Coupling Method
该文在一般状态空间下研究马氏链指数遍历性,指数遍历马氏链,增加条件π(fp)<∞,p>1,利用耦合方法得到了存在满的吸收集,使得马氏链在其上是f-指数遍历的.
We investigate the ergodic property of Markov chains in general state space, an exponentially ergodic conclusion Markov chains, adding conditions π(fp)<∞,p>1. By using the coupling method, there exists the full absorption set, such that the Markov chain is f-exponentially ergodic on it.
马氏链 / 耦合 / f-范数 / 指数遍历 {{custom_keyword}} /
Markov chains / coupling / f-norm / exponential ergodic {{custom_keyword}} /
[1] Ahmed M.A., Alkhamis T. M., Simulation–based optimization using simulated annealing with ranking and selection, Computers Operations Research, 2002, 29(4): 387–402.
[2] Andradottir S., A method for discrete stochastic optimization, Management Science, 1995, 41(12): 1946– 1961.
[3] Bowerman B., David H.T., Isaacson D., The convergence of Cesaro averages for certain non-stationary Markov chains, Stochastic Processes and Their Applications, 1977, 5(3): 221–230.
[4] Chen M. F., From Markov Chains to Non-equilibrium Particle Systems (Second Edition), World Scientific, Singapore, 2004.
[5] Gutjahr W.J., Pflug G., Simulated annealing for noisy cost functions, Journal of Global Optimization, 1996, 8: l–13.
[6] Lindvall T., Lectures on the Coupling Method, Wiley, New York, 1992.
[7] Meyn S. P., Tweedie R. L., Markov Chains and Stochastic Stability (1st Edition), Springer-Verlag, London, 1993.
[8] Zhang S. Y., Regularity and existence of invariant measures for jump processes, Acta Math. Sin., Chin. Ser., 2005, 48(4): 785–788.
[9] Zhang S. Y., Existence of the optimal measurable coupling and ergodicity for markov processes, Science in China, Ser. A, 1999, 42: 58–67.
[10] Zhang S. Y., The measurable coupling and probability distance of transition probability, Chinese Annals of Mathematics, 1995, 16(6): 769–775.
[11] Zhang S. Y., Xu K., The existence of optimal measurable coupling of transition probability, Acta Math. Sin., Chin. Ser., 1997, 40(1): 5–13.
湖北工程学院科研基金资助项目(201506)
/
〈 |
|
〉 |