Elements of multivariate time series analysis. 2nd ed. (English) Zbl 0873.62086

Springer Series in Statistics. New York, NY: Springer. xvii, 357 p. (1997).
For the review of the first edition from 1993 of this book see Zbl 0783.62072. This new edition has in general the same level and structure, with several changes and revisions. Some additional topics have been added, in particular the following: (1) There is a new chapter on the so-called ARMAX models (linear models with exogeneous variables); (2) State-space forms of the ARMA model are included at an earlier stage; (3) There is a new appendix on the relationship between the AR and MA parameter coefficient matrices and the corresponding covariance matrices of a vector ARMA process; (4) There is a new section on the information criterion AIC; (5) There is a new section on a canonical analysis for vector AR processes.
Some additional multivariate data sets are included, and also a small number of additional exercises. An author index has been added. The total length of the book increased from 263 to 357 pages, and the references went from 9 to 13 pages.


62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62-01 Introductory exposition (textbooks, tutorial papers, etc.) pertaining to statistics
62M20 Inference from stochastic processes and prediction
62H12 Estimation in multivariate analysis