J4 ›› 2009, Vol. 47 ›› Issue (05): 871-876.

• 数学 • 上一篇    下一篇

平稳ARIMA(p,d,q)模型的经验似然推断

 张海祥, 王德辉   

  1. 吉林大学 数学学院, 长春 130012
  • 收稿日期:2008-11-20 出版日期:2009-09-26 发布日期:2009-11-03
  • 通讯作者: 王德辉 E-mail:Wangdehui69@163.com.

Empirical Likelihood Inference for the StationaryARIMA(p,d,q) Model

 ZHANG Hai-Xiang, WANG De-Hui   

  1. College of Mathematics, Jilin University, Changchun 130012, China
  • Received:2008-11-20 Online:2009-09-26 Published:2009-11-03
  • Contact: WANG De-Hui E-mail:Wangdehui69@163.com.

摘要:

研究平稳ARIMA(p,d,q)模型的经验似然方法, 利用周
期图纵坐标I(ωj)的极限性质, 给出了参数的估计方程及Whittle’s估计因子, 并证明了
对数截面经验似然比统计量L(β)的极限分布是自由度为p+q+2的χ2分布, 同时给出了模
型参数的置信域.

关键词: ARIMA模型; 估计方程; 经验似然; 周期图; 谱密度

Abstract:

he authors  studied the empirical likelihood method for
stationary ARIMA(p,d,q) model, which is based on the periodogram ordinate’s
 asymptotical property, obtained the estimation equation and whittle’s estimat
or for the model parameter and  the distribution of the log profile empiri
cal likelihood ratio statistic L(β) which is asymptotically distributed as
χ2p+q+2, as well as the confidence regions for the parameters.

Key words:  ARIMA model, estimation equation, empirical likelihood, periodogram, spectral density

中图分类号: 

  • O212.7