Bayesian probabilistic extensions of a deterministic model (Q1424596)

The paper presents a stochastic extension of Boolean regression within a Bayesian framework which implies the addition of a Bernoulli-like process to the deterministic model. Methods of model estimation and checking are discussed as well as their application to a real data example (on definition of human emotions). Advantages of such Bayesian approach include a proper account of the uncertainty in the model estimation and various possibilities for model checking (using posterior predictive checks).