数学理论与应用 ›› 2021, Vol. 41 ›› Issue (2): 83-.

• • 上一篇    下一篇

基于IGGⅢ权函数的稳健ARMA残差控制图构建

黎小丽 李时敏1,*   孔盈莹2   

  1. 1.广东财经大学统计与数学学院,广州,510320; 2. 广东财经大学会计学院,广州,510320
  • 出版日期:2021-06-30 发布日期:2021-08-18
  • 通讯作者: 李时敏(1983—),男,湖南岳阳人,博士,副教授,研究方向:经济统计分析;邮箱:lism1983@126.com
  • 基金资助:

    本文得到广东省自然科学基金面上项目(2019A1515011885)资助

Based on the Robust ARMA Residual Control Chart of IGGⅢ Weight Function

  1. 1.School of Mathematics and Statistics, Guangdong University of Finance and Economics, Guangzhou 510320, China; 2.School of Accounting, Guangdong University of Finance and Economics, Guangzhou 510320, China
  • Online:2021-06-30 Published:2021-08-18

摘要:

针对传统ARMA残差控制图易受异常值的影响, 本文首先用由中国科学院测量与地球物理研究所提出的抗差方法IGGⅢ权函数建立稳健ARMA模型, 得到独立同分布残差序列. 接着用加权三均值和平均绝对离差分别估计均值和标准差, 进而构建稳健ARMA残差控制图. 模拟与实证检验均表明:本文所构建的稳健控制图能更好地抵抗异常值干扰, 监控效果优于传统控制图.

关键词: 稳健估计 ,  异常值 , 稳健ARMA残差控制图 , IGGⅢ权函数

Abstract:


Since the classical ARMA residual control chart is susceptible to outlier values, this paper first establishes a robust ARMA model based on the Institute of Geodesy & Geophysics III  IGGIII weight function to obtains an independent residual distribution sequence. Then a robust residual control chart is constructed based on the estimations of the mean and standard deviation by the weighted three mean and average absolute deviation, respectively. Simulations and empirical tests show that the robust control chart can resist the outlier interferences better, and the monitoring effect is better than that of the traditional control chart. 

Key words: Robust estimation ,  Outliers ,  Robust ARMA residual control chart ,  IGGⅢ weight function