×

Climate time series analysis. Classical statistical and bootstrap methods. 2nd ed. (English) Zbl 1300.86001

Atmospheric and Oceanographic Sciences Library 51. Cham: Springer (ISBN 978-3-319-04449-1/hbk; 978-3-319-04450-7/ebook). xxxi, 454 p. (2014).
The first edition of this book was published in 2010. Now, the second edition has unchanged structure but improves and updates some procedures. The author presents a carefully written research monograph on a very complex topic: time series analysis of the climate system. Assumptions for applying classical statistical methods are not fulfilled very often, like nonormal distribution shape, uneven spacing, time scale uncertainties. Therefore the author prefers bootstrap algorithms to derive more realistic error bars. In many cases bootstrap confidence intervals are presented. The methods are carefully testet by Monte-Carlo experiments and applied to a lot of real data from different data archives. From the introduction: “This book presents the bootstrap approach to a number of statistical analysis methods that have been found useful for analysing climate time series at least by the author. Linear and nonlinear regression (Chapter 4), spectral analysis (Chapter 5) and extreme value time series analysis (Chapter 6) are explained in case of univariate climate time series analysis (Part II). Correlation (Chapter 7) as well as lagged and other variants of regression (Chapter 8) come from the field of bivariate time series (Part III). Application of each method is illustrated with one or more climate time series, several of which already presented.” Every chapter is completed by an informative section “Background Material” where alternative techniques and a look at the literature are presented. The book comes to an end with a very long list of references, with far more than 1000 entries.

MSC:

86-02 Research exposition (monographs, survey articles) pertaining to geophysics
86A32 Geostatistics
86A10 Meteorology and atmospheric physics
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62P12 Applications of statistics to environmental and related topics
62G09 Nonparametric statistical resampling methods
PDFBibTeX XMLCite
Full Text: DOI