Analysis of time series structure. SSA and related techniques (Q2712748)

This monograph is devoted to a novel technique of time series analysis which is often called 'singular-spectrum analysis' (SSA). This technique incorporates the elements of classical time series analysis, multivariate statistics, multivariate geometry, dynamical systems and signal processing. But strictly speaking, SSA is not a statistical method in terms of classical statistics, in particular typically no assumptions concerning either signal or noise while performing the analysis are done. SSA is a model-free technique and it is more an explanatory model-building tool with the aim of decomposition of the original series into a sum of a small number of interpretable components such as a slowly varying trend, oscillatory components and noise.NEWLINENEWLINENEWLINEIt is worth to notice that SSA can be used for analyzing relatively short series. SSA has become a relatively well-known tool in meteorology, climatology and geophysics and also in nonlinear physics and signal processing, although this technique is not widely known among statisticians and econometricians. The material of the book touches many different fields and a large part of it will be useful for a wide circle of readers who need or have an interest in time series analysis.