一类随机环境下非线性自回归条件异方差模型的遍历性  

A sufficient condition of the companion geometric ergodicity to a general nonlinear time series model

作　　者：宋允全[1] 渐令[1] 

机构地区：[1]中国石油大学(华东)数学科学与计算技术学院,山东东营257061

出　　处：《云南大学学报（自然科学版）》2008年第S2期133-138,共6页Journal of Yunnan University(Natural Sciences Edition)

摘　　要：用"补充变量"的马氏化方法和一般状态空间马氏链的基本理论来分析讨论一类一般的随机环境中的非线性时间序列模型χt+1=f1(χ1)+…+fm(χt-m+1)+tε+1(Zt+1)h1(tχ)…hm(χt-m+1),χ0,χ-1,…,χ1-m∈R.首先,得到与该模型相应的马氏链理论并提出其伴随遍历和伴随几何遍历的定义;其次,给出该模型伴随几何遍历的一个充分条件.Following the idea and method ahead,applying the Markovnization and the theories of general state space Markov Chain,it has studied one model with stochastic delay: χt+1=f1(χt)+…+fm(χt-m+1+εt+1(Zt+1)h1(χt)…hm(χt-m+1),χ0,χ-1,…,χ1-m∈R.It first,another multivariate nonlinear AR model mixed with an ARCH term is studied and corresponding MARKOV chain theory is got,and the definition of companion ergodicity of this model is given.At last,a sufficient condition of the companion geometric ergodicity to this general...

关 键 词：非线性自回归异方差 伴随几何遍历性 μm×λ-不可约 马氏化 

分 类 号：O212.1[理学—概率论与数理统计]