随机有序马尔可夫链收敛速度显式界的注记  

Notes on Explicit Bounds for Convergence Rates of Stochastically Ordered Markov Chains

作　　者：符方健[1] 

机构地区：[1]琼台师范高等专科学校,海南海口571100

出　　处：《数学的实践与认识》2010年第20期206-212,共7页Mathematics in Practice and Theory

基　　金：国家自然科学基金(60574002);海南省自然科学基金(808250)

摘　　要：弱化Scott与Tweedie在计算马氏链收敛速度界时的条件,即变一步转移概率为m(m≥1)步转移概率,并运用不同于Scott与Tweedie的方法,计算出马氏链几何收敛速度r^n的界,从而推广了已有的结论.Scott and Tweedie find condition under which there are bounds on the conve＇rgence rates of Markov chains.This paper is an attempt to weaken that condition,that is,I manage to change one-step transition probability into m-step transition probability.At the same time,by using the different means from Scott and Tweedie,I compute the bounds for geometric convergence rates r^n of Markov chains,and therefore generalize the existent results.

关 键 词：马尔可夫链 m-骨架 漂移不等式 f-遍历 收敛速度 

分 类 号：O211.62[理学—概率论与数理统计]