Bayesian Modeling of Poisson Time Series
Tevfik Aktekin · 2011
In this study, we introduce a general class of Poisson time series models. In doing so, we develop discrete time state space models with Poisson measurements and their Bayesian inference via Markov chain Monte Carlo methods. Furthermore, we discuss issues of model properties, sequential updating/filtering, smoothing and forecasting. In order to show the implementation of the proposed models, we use real count data and discuss further implications of the Bayesian approach. ∗Refik Soyer’s research was partially supported by the National Science Foundation under Grant DMS-0915156 with the George Washington University.