J4 ›› 2011, Vol. 49 ›› Issue (04): 615-624.

• 数学 • 上一篇    下一篇

MA(∞)误差下部分线性模型的经验似然统计推断

于卓熙1, 王德辉2   

  1. 1. 吉林财经大学 管理科学与信息工程学院, 长春 130117|2. 吉林大学 数学学院, 长春 130012
  • 收稿日期:2010-11-17 出版日期:2011-07-26 发布日期:2011-08-16
  • 通讯作者: 王德辉 E-mail:Wangdehui69@163.com

Empirical Likelihood in Partial Linear Modelswith MA(∞) Error Process

YU Zhuoxi1, WANG Dehui2   

  1. 1. School of Management Science and Information Engineering, Jilin University of Finance and Economics, Changchun 130117, China|2. College of Mathematics, Jilin University, Changchun 130012, China
  • Received:2010-11-17 Online:2011-07-26 Published:2011-08-16
  • Contact: WANG Dehui E-mail:Wangdehui69@163.com

摘要:

应用经验似然方法, 针对误差为不可观测无穷阶滑动平均过程的部分线性模型, 构造了回归参数的对数经验似然比检验统计量, 并证明了统计量在参数取真值时渐近地服从χ2分布, 构造了参数的置信区间. 模拟计算表明, 经验似然方法优于最小二乘方法.

关键词: 部分线性模型, MA(∞)误差过程, 经验似然

Abstract:

The authors concerned  the partial linear models with serially correlated random errors which are not observed and modeled by a moving\|average process of infinte order. We proposed an empirical loglikelihood ratio statistic for the regression coefficients. Our results show that the statistic is asymptotically chisquare distributed and the corresponding confidence interval can be constructed accordingly. A simulation illustrates that the empirical likelihood method works better than the ordinary least squares method.

Key words: partial linear regression model, MA(∞) error process, empirical likelihood

中图分类号: 

  • O212.7