样本相对熵率与一类关于随机选择系统的强偏差定理  

Sample Relative Entropy Rate and a Class of Strong Deviation Theorems on Random Select System

作　　者：赵静[1] 魏杰[2] 孙利民[2] 

机构地区：[1]天津工业大学理学院,天津300160 [2]军事交通学院基础部,天津300161

出　　处：《河北工业大学学报》2005年第2期110-114,共5页Journal of Hebei University of Technology

摘　　要：将样本相对熵率和随机选择系统的概念引入任意整值随机变量序列的研究,在样本空间的子集上把样本相对熵率作为任意整值随机变量序列与非齐次马氏链之间偏差的一种度量,给出了一类关于随机选择系统的强偏差定理.The notion of sample relative entropy rate and random select system are introduced to investigate arbitrary in- teger-valued random variables, and by use of sample relative entropy rate as a measure of the difference between the se- quences of arbitrary integer-valued random variables and nonhomogeneous Markov chains, considered on certain sets of the sample space, a class of strong deviation theorems on random select system are obtained.

关 键 词：样本相对熵率 随机选择系统 非齐次马氏链 强偏差定理 

分 类 号：O211.4[理学—概率论与数理统计]