Modeling time series data is of paramount importance considering that the natural phenomena that we experience are actually time series, that is, sequences of observations that can be realized in chronological order. Of particular interest is the modeling and prediction of hydro-meteorological time series data considering the influence on humans, and this is what this book, spread over seven chapters, is all about.
Chapter 1 deals with “hypotheses testing on meteorological time series.” This includes tests on normality, homoscedasticity, autocorrelation, outlier detection, and goodness of fit. Chapter 2 targets “mathematical methods applied for hydro-meteorological time series modeling.” The topics covered include the classical decomposition method, the Box-Jenkins approach and stationarity tests, support vector regression (SVR), genetic algorithms, general regression neural network (GRNN), and wavelets. Chapter 3 contains “models for precipitation series” like autoregressive moving average (ARMA), SVR, GRNN, hybrid models, decomposition, and wavelet models, to name a few. Chapter 4 covers “modeling the precipitation evolution at [a] regional scale.” Chapter 5 focuses on “analysis and models for surface water quality.” Chapter 6 is on “models for pollutants dissipation,” while the last chapter (7) covers “spatial interpolation with applications.” All this is supported by a comprehensive and helpful bibliography.
I enjoyed reading this carefully written book and would certainly recommend it to postgraduate students and researchers of meteorology, and applied mathematicians and statisticians who deal with such environmental science data.