English

应用科学学报

• 论文 • 上一篇    下一篇

纵向数据线性混合效应模型的统计分析

孙 燕1,2 柴根象1   

  1. 1. 同济大学 应用数学系,上海 200092；

    2. 上海财经大学 经济学院数量经济研究所 上海 200433

  • 收稿日期:2004-10-12 修回日期:2005-04-18 出版日期:2006-01-31 发布日期:2006-01-31

Statistical Analysis of Linear Mixed Model for Longitudinal Data

SUN Yan1,2 CHAI Gen-xiang1   

  1. 1. Department of Applied Mathematics,Tongji Unversity,Shanghai 20092,China
    2. School of Economics, Shanghai Unversity of Finance and Economics,Shanghai 200433,China
  • Received:2004-10-12 Revised:2005-04-18 Online:2006-01-31 Published:2006-01-31

被引次数

7

摘要： 考虑纵向数据的一般线性混合效应模型，在不设定误差和随机效应分布类型已知的情况下，给出了随机效应本身，回归系数和误差方程的估计量，并在一般设计点列和一般矩条件下，建立了估计量的强相合性和渐近正态性.

关键词: 纵向数据, 随机效应, 混合效应, 强相合性, 渐近正态性.

Abstract: The linear mixed model for longitudinal data proposed by Laird and Ware is studied.Estimators of the random effects, regression coeddicients and error variance are established without assuming any distributions of random effects and random errors.Strong consistency and ssymptotic normality are discussed under the general conditions of design points and moments.

Key words: longitudinal data, random effects, mixed effects, strong consistency, asymptotic normality