zbMATH — the first resource for mathematics

Predictions in time series using regression models. (English) Zbl 1011.62102
New York, NY: Springer. ix, 231 p. (2002).
The models proposed by the author concern the aspects of time series which can be modelled by linear and nonlinear regression. The author thinks on applications of his time series methods in the fields of ecology, econometrics and finance. Often in these fields a detailed description of a time varying mean and covariance structure is required.
The book consists of five chapters with literature references and a subject index. In the first chapter, results and examples of Hilbert spaces are given which are related to statistics and are used later on in the book. In the next chapters, basic results of random processes, spectral theory and especially of time series are summarized. With the chapters on estimation and prediction the author presents the main subject of methods and models announced in the book title. A lot of numerical examples are given.
Although in the book many general concepts and results for time series other than regression related are presented, it may not be recommended as a short introduction to time series, rather as a strong mathematical exposition.

62M20 Inference from stochastic processes and prediction
62-02 Research exposition (monographs, survey articles) pertaining to statistics