ABSTRACT
This chapter introduces various linear models. Linear dynamic models are now being studied by statisticians, but in the past have received the most attention of engineers who are interested in communication theory, navigation systems, and tracking of satellites. This class of models should prove to be very useful in the statistical analysis of time series. The regression and design models are quite similar and each is expressed in terms of general linear model. The term "structural change" is used by econometricians to denote a change in the parameters of a model which explains a relationship between economic variables. The Kalman filter is an algorithm which recursively computes the mean of the posterior distribution of a parameter of a dynamic model. This chapter focuses on the history of Bayesian inference, the subjective interpretation of probability, the main ingredients of a Bayesian analysis, the various types of prior information, the implementation of prior information, and the advantages and disadvantages of Bayesian inference.
