The First Step in Bayesian Time Series-- Linear Regression

Today time series forecasting is ubiquitous, and decision-making processes in companies depend heavily on their ability to predict the future. Through a short series of articles I will present you with a possible approach to this kind of problems, combining state-space models with Bayesian statistics. In the initial articles, I will take some of the examples from the book An Introduction to State Space Time Series Analysis from Jacques J.F. Commandeur and Siem Jan Koopman [1]. It comprises a well-known introduction to the subject of state-space modeling applied to the time series domain. In classical regression analysis, it is assumed a linear relationship between a dependent variable y and a predictor variable x.