Approximations to the Distributions of the Likelihood Ratio Statistics for Testing the Hypotheses on Covariance Matrices and Mean Vectors Simultaneously,

摘要

In this report, the authors approximated the distributions of certain powers of the likelihood ratio statistics for testing the hypothesis of the homogeneity of populations and the hypothesis that the covariance matrix and mean vector are equal to specified values when the underlying populations are real or complex multivariate normal. The accuracy of these approximations is sufficient for practical purposes.