Multivariate Analysis -- III contains the proceedings of the Third International Symposium on Multivariate Analysis held at Wright State University in Dayton, Ohio, on June 19-24, 1972. The papers explore the theory and applications of multivariate analysis and cover areas such as time series and stochastic processes; distribution theory and inference; characteristic functions and characterizations; and design and analysis of experiments. Classification, modeling, and reliability are also discussed. Comprised of 27 chapters, this volume begins with an introduction to two-dimensional random fields, giving results for a class of Gaussian processes with a multidimensional time parameter. The next chapter deals with concepts of consistency in spectral estimation for multivariate time series and considers the alternative of estimating the spectral distribution function or the spectral density function. Abstract martingales and ergodic theory are also examined, along with methods for assessing multivariate normality; inference and redundant parameters; characterization of the multivariate geometric distribution; and max-min designs in the analysis of variance. This monograph will be useful to statisticians and probabilists, as well as to scientists in other disciplines who are broadly interested in multivariate analysis.