Bayesian Smoothing of Dynamic linear model – Step model

Abstract

Smoothing is a kind of estimation of a parameter or a state variable. It is also called data smoothing. In this paper, the Bayesian smoothing for the dynamic linear model (step model) has been considered. The use of Bayesian smoothing means the reconsideration of new data for better enhancement and error reduction. Implementing any new data in the estimation process is a type of backward filtering to get a more precise estimation of the model parameters (state variable). In this paper, the mathematical content is highlighted to obtain a practical application as soon as real data are provided