Regression methods have been an integral part of time series analysis for over a century. Recently, new developments have made major strides in such areas as non–continuous data where a linear model is not appropriate. This book introduces the reader to newer developments and more diverse regression models and methods for time series analysis.
Accessible to anyone who is familiar with the basic modern concepts of statistical inference, Regression Models for Time Series Analysis provides a much–needed examination of recent statistical developments. Primary among them is the important class of models known as generalized linear models (GLM) which provides, under some conditions, a unified regression theory suitable for continuous, categorical, and count data.
The authors extend GLM methodology systematically to time series where the primary and covariate data are both random and stochastically dependent. They introduce readers to various regression models developed during the last thirty years or so and summarize classical and more recent results concerning state space models. To conclude, they present a Bayesian approach to prediction and interpolation in spatial data adapted to time series that may be short and/or observed irregularly. Real data applications and further results are presented throughout by means of chapter problems and complements.
Notably, the book covers:
∗ Important recent developments in Kalman filtering, dynamic GLMs, and state–space modeling
∗ Associated computational issues such as Markov chain, Monte Carlo, and the EM–algorithm
∗ Prediction and interpolation
∗ Stationary processes
Times Series Following Generalized Linear Models.
Regression Models for Binary Time Series.
Regression Models for Categorical Time Series.
Regression Models for Count Time Series.
Other Models and Alternative Approaches.
State Space Models.
Prediction and Interpolation.
Appendix: Elements of Stationary Processes.