Aimed at professional statisticians as well as graduate students, this text presents an introduction to multivariate statistical analysis. Anderson (emeritus, statistics and economics, Stanford U.) discusses such topics as the estimation of the mean vector and the covariance matrix, the classification of observations, and the distributions of characteristic roots and vectors. Updated to reflect recent advances in the field (the first edition was published in 1957), the third edition features expanded information on a number of topics, including a new chapter on patterns of dependence and graphical models. Annotation (c) Book News, Inc., Portland, OR (booknews.com)