Bayesian Information Criterion (BIC), is an order estimation criterion used in the Box-Jenkins specification stage of model building. A model that minimized the BIC is considered to be the most appropriate model. When several models may be appropriate, choose the one with the smallest BIC. The AIC and BIC differ in their second terms, which are penalty functions for extra parameters. Often AIC and BIC will lead to the same model choice.

Back to Forecasting Methods