Regression models for time series analysis /

Regression models have been used often in forecasting and decomposition of time series for many years. Previous work in this area has discussed linear models for time series, assuming continuity associated with Slutski and Yule in the 1920s, Wold in the 1930s, and Box and Jenkins in the 1970s. Kedem (Univ. of Maryland) and Fokianos (Univ. of Cyprus) examine recent statistical developments where noncontinuous data are analyzed and where linear models are not appropriate. One recently cited development is a class of models known as generalized linear models (GLM), originally cited by J.A. Nelder and R.W.M. Wedderburn in a 1970s paper, which has applications for continuous, categorical, and count data. Also included are recent developments in dynamic GLM, state-space modeling, and Kalman filtering. Major topics covered include time series following GLM and regression models for binary time series, categorical time series, and count time series. A master's degree-level background in statistical inference and applied stochastic processes is essential for understanding the material in this book. Highly recommended for graduate-level library collections. Graduate students; faculty. D. J. Gougeon University of Scranton