ABSTRACT
All the models introduced so far and many other linear time series models can be represented in a form called state space form. The great advantage of representing time series models in the state space form is the availability of a set of general algorithms, the Kalman filter and its relatives, for the estimation by linear prediction of the unobservable components and for the computation of the Gaussian likelihood, which can be numerically maximised to obtain maximum likelihood (ML) estimate of the model’s parameters.
