A Bayesian Model about Relationship between Weather and Sales of Medicine for Cold

摘要

Weather factors affecting the oscillation of the sales of cold medicine, may be various, uncertain and intermittent. A Bayesian model is built in this paper, to analyze the relationship between the weather factors and the sales of cold medicine, and BDE(Bayesian Dirichlet equivalent) score function is applied for learning Bayesian network structure so that the predominant weather factors could be identified. Some evaluation indicators are used to evaluate a reliable predictive model. Numerical examples confirm the feasibility and the reliability of the derived Bayesian model.