State Space Modeling of Time Series

Adopting a state space approach to time series modelling, this volume aims to present a new, computer-orientated method for building models for vector-valued time series.

---

The author adopts a state space approach to time series modelling in this volume to provide a new, computer-orientated method for building models for vector-valued time series. Background material leading up to the two types of estimators of the state space models is collected and presented coherently in four consecutive chapters. This edition has been revised to provide more comprehensive descriptions of state space models for autoregressive models commonly used in the econometric and statistical literature. Backward innovation models are introduced in addition to the forward innovation models, and both are used to construct instrumental variable estimators for the model matrices. Further items in this edition include statistical properties of these two types of estimators, more details on multiplier analysis and the identification of structural models using estimated models, incorporation of exogenous signals and choice of model size. A chapter is devoted to the modelling of integrated, nearly integrated and co-integrated time series.