数学物理学报(英文版) ›› 1994, Vol. 14 ›› Issue (S1): 103-109.

• 论文 • 上一篇    下一篇

THE RELATIVE EFFICIENCIES OF LEAST SQUARES IN LINEAR MODELS

陈建宝1, 詹金龙2   

  1. 1. Institute of Applied Mathematics, Yunnan Province, Kunming 650091, China;
    2. Department of Basic Sciences, Kunming Institute of Technology, Kunming 650093, China
  • 收稿日期:1992-11-05 出版日期:1994-12-31 发布日期:1994-12-31

THE RELATIVE EFFICIENCIES OF LEAST SQUARES IN LINEAR MODELS

Chen Jianbao1, Zhan Jinlong2   

  1. 1. Institute of Applied Mathematics, Yunnan Province, Kunming 650091, China;
    2. Department of Basic Sciences, Kunming Institute of Technology, Kunming 650093, China
  • Received:1992-11-05 Online:1994-12-31 Published:1994-12-31

摘要: The present paper discusses the relative efficiencies of the least square estimates in linear models. For Gauss-Markoff model:Y=Xe + e,E(e)=0, Cov(e)=σ2V, an new efficiency of least square estimate for linearly estimable function c' τ is proposed and its lower bound is given. For variance component model:Y=+ e, E(e)=0, Cov(e)=∑i=1mσ2V, an new efficiency of least square estimate for linearly estimable function c' τ is introduced for the first time and its lower bound, which is independent of unknown parameters, is also obtained.

关键词: Linear model, Gauss-Markov, Variance component, LSE, BLUE, Efficiency

Abstract: The present paper discusses the relative efficiencies of the least square estimates in linear models. For Gauss-Markoff model:Y=Xe + e,E(e)=0, Cov(e)=σ2V, an new efficiency of least square estimate for linearly estimable function c' τ is proposed and its lower bound is given. For variance component model:Y=+ e, E(e)=0, Cov(e)=∑i=1mσ2V, an new efficiency of least square estimate for linearly estimable function c' τ is introduced for the first time and its lower bound, which is independent of unknown parameters, is also obtained.

Key words: Linear model, Gauss-Markov, Variance component, LSE, BLUE, Efficiency