论文信息 - MANOVA in the multivariate components of variance model

MANOVA in the multivariate components of variance model

Conditions are obtained for the multivariate components of variance model to admit a multivariate analysis of variance (MANOVA). MANOVA in this setup is defined as a partition of the sum of squares and sum of products matrix into independent Wishart matrices. A minimal sufficient statistic is exhibited using the terms in the MANOVA and its completeness is discussed. The results are illustrated using examples.

Thomas Mathew | T. Mathew

[1] Yasuo Amemiya,et al. What Should be Done When an Estimated between-Group Covariance Matrix is not Nonnegative Definite? , 1985 .

[2] A. Albert. When is a Sum of Squares an Analysis of Variance , 1976 .

[3] Steven F. Arnold,et al. Linear Models with Exchangeably Distributed Errors , 1979 .

[4] Calyampudi R. Rao,et al. LIKELIHOOD RATIO TESTS FOR RELATIONSHIPS BETWEEN TWO COVARIANCE MATRICES , 1982 .

[5] T. W. Anderson. Estimating Linear Statistical Relationships , 1984 .

[6] K. G. Brown. Sub-Balanced Data and the Mixed Analysis of Variance , 1983 .

[7] Ingram Olkin,et al. Maximum Likelihood Estimators and Likelihood Ratio Criteria in Multivariate Components of Variance , 1986 .

[8] On Analysis of Variance in the Mixed Model , 1984 .

[9] F. Graybill,et al. Theorems Concerning Eisenhart's Model II , 1961 .

[10] P. Krishnaiah,et al. On Covariance Structures. , 1975 .

[11] James R. Schott,et al. A multivariate one-way classification model with random effects , 1984 .

[12] J. Miller,et al. Explicit Maximum Likelihood Estimates from Balanced Data in the Mixed Model of the Analysis of Variance , 1980 .

[13] Roman Zmyślony,et al. Completeness for a family of normal distributions , 1980 .

[14] Terence P. Speed. What is an Analysis of Variance , 1987 .