Heuristic Analysis of Time Series Internal Structure

A method of analysis of Time Series Internal Structures based on Singular Spectrum Analysis is discussed. It has been shown that in the case when the Time Series contains deterministic additive components rank of the trajectory matrices equal to number of parameters of the components. Also it was proved that both eigen and factor vectors repeat shapes of the additive components and both eigen values and eigen vectors can be divided into additive groups. Some useful patterns of deterministic components were identified, which permit to provide graphical analysis of times series Internal Structures.