Two-level Mean and Covariance Structures: Maximum Likelihood via an EM Algorithm

Two-level Mean and Covariance Structures: Maximum Likelihood via an EM Algorithm ∗ Peter M. Bentler † and Jiajuan Liang University of California, Los Angeles Department of Psychology, Box 951563 Los Angeles, CA 90095-1563, U.S.A. Abstract An EM-type gradient algorithm for analysis of maximum likelihood es- timation of the two-level structural equation model with both mean and covariance structures is proposed. The model considered in this paper is a generalization of that studied by Lee and Poon (1998). Approximate standard error of the maximum likelihood estimation and the chi-squared statistic for testing the model fit are given. Simulation studies show that the proposed EM gradient algorithm converges very fast and need not be accelerated by existing techniques in the literature. Key Words: EM gradient algorithm, maximum likelihood estimation, mean and covariance structures, multilevel structural equation modeling, multivariate normal distribution This work was supported by National Institute on Drug Abuse grants DA01070 and DA00017 The correspondence author, E-mail: bentler@ucla.edu

[1]  Wai-Yin Poon,et al.  ANALYSIS OF TWO-LEVEL STRUCTURAL EQUATION MODELS VIA EM TYPE ALGORITHMS , 1998 .

[2]  Stephen W. Raudenbush,et al.  Maximum likelihood estimation for unbalanced multilevel covariance structure models via the EM algorithm , 1995 .

[3]  Terry E. Duncan,et al.  Latent Variable Modeling of Longitudinal and Multilevel Substance Use Data. , 1997, Multivariate behavioral research.

[4]  Sik-Yum Lee,et al.  Multilevel analysis of structural equation models , 1990 .

[5]  N. L. Johnson,et al.  Linear Statistical Inference and Its Applications , 1966 .

[6]  B. Muthén,et al.  Multilevel Covariance Structure Analysis , 1994 .

[7]  P. Bentler,et al.  ML Estimation of Mean and Covariance Structures with Missing Data Using Complete Data Routines , 1999 .

[8]  Bengt Muthen,et al.  10. Latent Variable Modeling of Longitudinal and Multilevel Data , 1997 .

[9]  Peter M. Bentler,et al.  EQS : structural equations program manual , 1989 .

[10]  Sik-Yum Lee,et al.  ANALYSIS OF COVARIANCE STRUCTURES WITH INDEPENDENT AND NON-IDENTICALLY DISTRIBUTED OBSERVATIONS , 1998 .

[11]  Roderick P. McDonald,et al.  A general model for two-level data with responses missing at random , 1993 .

[12]  K. Lange.  A gradient algorithm locally equivalent to the EM algorithm , 1995 .

[13]  R. Jennrich,et al.  Acceleration of the EM Algorithm by using Quasi‐Newton Methods , 1997 .

[14]  Risto Lehtonen,et al.  Multilevel Statistical Models , 2005 .

[15]  Harvey Goldstein,et al.  Multilevel Models in Education and Social Research , 1987 .

[16]  B. Hoadley.  Asymptotic Properties of Maximum Likelihood Estimators for the Independent Not Identically Distributed Case , 1971 .

[17]  D. Harville.  Matrix Algebra From a Statistician's Perspective , 1998 .

[18]  R. Darrell Bock,et al.  Multilevel analysis of educational data , 1989 .

[19]  H. Goldstein,et al.  Multilevel Models in Educational and Social Research. , 1989 .

[20]  New York Dover,et al.  ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM , 1983 .

[21]  J. Kalbfleisch,et al.  Between- and within-cluster covariate effects in the analysis of clustered data. , 1998, Biometrics.

[22]  Risto Lethonen.  Multilevel Statistical Models (3rd ed.) , 2005 .

[23]  Harvey Goldstein,et al.  A general model for the analysis of multilevel data , 1988 .