MARKOV-PROCESSES AND EXPONENTIAL-FAMILIES

被引：1
|
作者：
YCART, B
机构：
[1] Laboratoire de Mathématiques Appliquées, Faculté des Sciences, Université de Pau, Pau
来源：
STOCHASTIC PROCESSES AND THEIR APPLICATIONS | 1992年 / 41卷 / 02期
关键词：
MARKOV PROCESSES; EXPONENTIAL FAMILIES;
D O I：
10.1016/0304-4149(92)90121-6
中图分类号：
O21 [概率论与数理统计]; C8 [统计学];
学科分类号：
020208 ; 070103 ; 0714 ;
摘要：
We study Markov processes the distribution of which stays for some interval of time in a given exponential family of distributions with one parameter. We show that the generator of such a process can be canonically associated to the generator of a Markov process on R, having the same stability property with respect to a natural exponential family of distributions on R. Examples of spin systems on a finite set and of Brownian motions on R(n) illustrate the results.
引用
收藏
页码：203 / 214
页数：12
相关论文
共 50 条
12345