Bayesian dynamic linear models (DLM) are useful in time series modelling because of the flexibility that they present in obtaining a good forecast. They are based on a decomposition of the relevant factors which explain the behavior of the series through a series of state parameters....