中文

Journal of Applied Sciences

• Articles • Previous Articles     Next Articles

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

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