随时间变化系数参数AR模型的Bayes估计

J4 ›› 2011, Vol. 49 ›› Issue (05): 857-860.

随时间变化系数参数AR模型的Bayes估计

施三支¹, 王德辉², 宋立新³, 闫丽¹

1. 长春理工大学理学院, 长春 130022|2. 吉林大学数学学院, 长春 130012;3. 大连理工大学数学学院, 辽宁大连 116024

收稿日期:2011-01-27 出版日期:2011-09-26 发布日期:2011-09-27
通讯作者: 王德辉 E-mail:wangdehui@jlu.edu.cn

Bayesian Estimation of Time Vary Parameter Autoregression Model

SHI Sanzhi¹, WANG Dehui², SONG Lixin³, YAN Li¹

1. School of Science, Changchun University of Science and Technology, Changchun 130022, China;2. College of Mathematics, Jilin |University, Changchun 130012, China;3. College of Mathematics, Dalian University of Technology, Dalian 116024, Liaoning Province, China

Received:2011-01-27 Online:2011-09-26 Published:2011-09-27
Contact: WANG Dehui E-mail:wangdehui@jlu.edu.cn

摘要/Abstract

摘要：

采用Bayes方法, 给出一维随时间变化系数自回归时间序列模型(TVPAR模型)中各参数的Bayes估计. 选取未知参数的先验分布为均匀分布和逆伽马分布, 采用MCEM算法, 给出了模型中各参数的估计算法, 并对模型进行了模拟计算与实证分析.

关键词: TVPAR模型； MCEM算法； Bayes估计；先验分布

Abstract:

The authors proposed the Bayesian method to estimate the parameters in a special time vary parameter autoregression model(TVPAR model). Assuming that coefficient parameters are uniform, and variance is inverse gamma distribution, we got the iterate steps of all parameters by means of MCEM algorithm. Simulation and a real data analysis were performed to illustrate the proposed estimation method.

Key words: TVPAR model, MCEM algorithm, Bayesian estimation, prior distribution

中图分类号:

O212.1

施三支, 王德辉, 宋立新, 闫丽. 随时间变化系数参数AR模型的Bayes估计[J]. J4, 2011, 49(05): 857-860.

SHI San-Qi, WANG De-Hui, SONG Li-Xin, YAN Li. Bayesian Estimation of Time Vary Parameter Autoregression Model[J]. J4, 2011, 49(05): 857-860.

[1]	史海芳, 姬永刚. 简单半序约束下多个正态总体分布参数的Bayes估计与等值检验[J]. J4, 2013, 51(01): 1-8.
[2]	齐芯, 王德辉. 序约束下AR(2)模型参数的拟似然估计[J]. J4, 2011, 49(03): 447-451.
[3]	王纯杰, 宋立新. 基于Copula函数的风险度量和保费定价模式[J]. J4, 2010, 48(02): 169-176.
[4]	朱复康, 李琦. INGARCH(1,1)模型参数的矩估计和Bayes估计[J]. J4, 2009, 47(05): 899-902.