Dynamic causal modeling

Bayesian model reduction

Bayesian model reduction is a method for computing the evidence and parameters of Bayesian models which differ only in the specification of their priors. Typically, a 'full' model is estimated using the available data using standard approaches. Then, hypotheses are tested by defining one or more 'reduced' models, which differ only in their priors. In the context of variational Bayes, the evidence and parameters of the reduced models can be computed analytically from the evidence and parameters of the full model. This has numerous applications, including scoring large numbers of models rapidly, and facilitating the estimation of hierarchical (Parametric Empirical Bayes) models.