Regression, Transfer Function and Noise Models
Polynomial and seasonal models provide two basic and very widely used classes of DLMs, both, of course, TSDLMs. A third important class is that based on regression relationships; the three classes combined via the principle of superposition provide for the vast majority of forms of behaviour encountered in practice for which linear models can provide adequate descriptions. Here we describe the theory and analysis of regression DLMs and important related models.
KeywordsTransfer Function State Vector Posterior Distribution Noise Model Noise Process
Unable to display preview. Download preview PDF.