时间序列趋势项的贝叶斯估计

journal6 ›› 2006, Vol. 27 ›› Issue (4): 22-24.

时间序列趋势项的贝叶斯估计

(1.吉首大学数学与计算机科学学院，湖南吉首 416000；2.吉首大学师范学院，湖南吉首 416000)

出版日期:2006-07-25 发布日期:2012-08-22
作者简介:欧祖军(1979-),男,湖南省宜章县人,吉首大学数学与计算机科学学院讲师,硕士,主要从事概率统计研究.
基金资助:
吉首大学校级科研项目(05JD046)

Bayesian Estimation of Time Series Model with Trend

(1.College of Mathematics and Computer Science，Jishou University,Jishou 416000，Hunan China;2.Normal College of Jishou University,Jishou 416000，Hunan China)

Online:2006-07-25 Published:2012-08-22

摘要/Abstract

摘要：系统地分析了AR(p)误差项的时间序列模型的数学模型及条件似然函数，并根据似然函数的统计结构构造了模型参数的共轭先验分布，研究了正态-混合Γ先验下模型的贝叶斯推断理论，包括趋势项的核估计参数及先验参数的后验分布的统计推断.

关键词： 贝叶斯估计, AR(p)误差, Gibbs抽样

Abstract: This paper systematically analyzes the mathematic model and conditional likelihood function of time series model with AR(p) error,and constructs conjugate prior distribution of parameters on the basis of the statistical structure of the function.In the condition of normal-mixed Γ prior distribution，the paper studies the Bayesian inference of the model,including the statistical inference of the parameter of the kernel estimation of trend and the prior parameters of posterior distribution.

Key words: Bayesian estimation, AR(p) error, Gibbs sampling

欧祖军, 李洪毅, 罗先发. 时间序列趋势项的贝叶斯估计[J]. journal6, 2006, 27(4): 22-24.

OU Zu-Jun , LI Hong-Yi, LUO Xian-Fa. Bayesian Estimation of Time Series Model with Trend[J]. journal6, 2006, 27(4): 22-24.