Bayesian Forecasting for Automobile Parts using Stochastic Simulation

This article presents the development and application of a simulation model that was used to forecast the demand of automobile parts using information from a car dealer in Mexico, D. F. In particular, this work illustrates, using a simple model, how stochastic simulation and Bayesian statistics can be combined to model and solve complex forecasting problems. The proposed framework is general enough to be applied to very detailed models of the system under study. The results obtained demonstrate how uncertainty on the parameters of the model can be incorporated, and the application using real data shows how a large sample size produces a posterior distribution that has little influence from the prior distribution