Maximum Likelihood Estimation of a Set of Covariance Matrices Under Lowner Order Restrictions with Applications to Balanced Multivariate Variance Components Models

The problem of maximum likelihood estimation of Lowner ordered covariance matrices is considered. It is shown that a dual formulation of this problem is tractable and important in its own right. The interplay between the primal and dual problems suggests a general algorithm for computing the solutions to these problems. This algorithm has application to some estimation problems in balanced multivariate variance components models. The speed of convergence is also discussed for the variance components models.

[1]  D. Harville,et al.  Computational aspects of likelihood-based inference for variance components. , 1990 .

[2]  Joseph Sedransk,et al.  Bayesian and frequentist predictive inference for the patterns of care studies , 1991 .

[3]  W. A. Thompson.  The Problem of Negative Estimates of Variance Components , 1962 .

[4]  S. R. Searle,et al.  The theory of linear models and multivariate analysis , 1981 .

[5]  W. N. Venables,et al.  Estimation of Variance Components and Applications. , 1990 .

[6]  C. R. Henderson.  ESTIMATION OF VARIANCE AND COVARIANCE COMPONENTS , 1953 .

[7]  J. Klotz,et al.  Maximum Likelihood Estimation of Multivariate Covariance Components for the Balanced One-Way Layout , 1969 .

[8]  Calyampudi Radhakrishna Rao,et al.  Linear Statistical Inference and its Applications , 1967 .

[9]  Karl Löwner.  Über monotone Matrixfunktionen , 1934 .

[10]  S. R. Searle.  Linear Models , 1971 .

[11]  L. Herbach.  Properties of Model II--Type Analysis of Variance Tests, A: Optimum Nature of the $F$-Test for Model II in the Balanced Case , 1959 .

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

[13]  R. Dykstra,et al.  Duality of I Projections and Maximum Likelihood Estimation for Log-Linear Models Under Cone Constraints , 1988 .

[14]  F. Pukelsheim.  Linear models and convex geometry: aspects of non-negative variance estimation 1 , 1981 .

[15]  F. Pukelsheim,et al.  On the duality between locally optimal tests and optimal experimental designs , 1985 .

[16]  Anil K. Bhargava,et al.  Exact probabilities of obtaining estimated non-positive definite between-group covariance matrices , 1982 .

[17]  D. Harville.  Maximum Likelihood Approaches to Variance Component Estimation and to Related Problems , 1977 .

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

[19]  R. H. Farrell,et al.  Multivariate calculation : use of the continuous groups , 1986 .

[20]  P. S. Dwyer.  Some Applications of Matrix Derivatives in Multivariate Analysis , 1967 .

[21]  C. Radhakrishna Rao,et al.  Minimum variance quadratic unbiased estimation of variance components , 1971 .

[22]  W. G. Hill,et al.  Probabilities of Non-Positive Definite between-Group or Genetic Covariance Matrices , 1978 .

[23]  H. D. Patterson,et al.  Recovery of inter-block information when block sizes are unequal , 1971 .

[24]  H. D. Brunk,et al.  The Isotonic Regression Problem and its Dual , 1972 .

[25]  F. T. Wright,et al.  Order restricted statistical inference , 1988 .

[26]  D. Luenberger.  Optimization by Vector Space Methods , 1968 .