Deviance Information Criterion

In statistics, the deviation information criterion (English deviance information criterion, DIC ) is a measure ( criterion ) for the prediction error of a model.

This measure is an information criterion and heard in the context of Bayesian method for model comparisons. The smaller the DIC, the better the model fit. The DIC can be seen as Bayesian equivalent of the AIC.

Very formulated roughly in the assessment of two models with different DIC applies: If there are differences greater than 10, the model with the higher DIC is definitely worse, differences between 5 and 10 are substantial, with differences less than 5 and distinctly different model formulations, it may be necessary to take both models into account.