The construction of stochastic dynamic models from empirical time series is practiced in a variety of disciplines, including engineering, ecology, and applied statistics, with specific forecasting aims. However, there have been few systematic expositions of the major problems facing model builders: determination of the plausible classes of models for the given time series by inspection of the series and examination of its characteristics, detailed comparison of the various classes of models, the role and methods of model validation, etc. Of course, there have been a number of books discussing some techniques for developing models for time series data, as mentioned in the text.


Read more at http://ebookee.org/Dynamic-Stochastic-Models-from-Empirical-Data_1057085.html#HO9oTsJIpGMhrHfj.99