ABSTRACT
A general class of models, arousing much interest in many areas of application, is that of state-space models. State-space models have also found increasing use in many types of time-series problems, including parameter estimation, smoothing and prediction. This chapter introduces state-space models for the time-series analyst, as well as describing the Kalman filter, which is an important general method of handling state-space models. The state variables are typically model parameters of some sort, such as regression coefficients in a regression model or parameters describing the state of a system in a rather different way. The application of state-space models to engineering problems, such as controlling a dynamic system, is fairly clear. An important difference between state-space modelling in time-series applications and in some engineering problems is that the structure and properties of a time series will usually not be assumed known a priori.
