A hierarchical Bayesian learning framework for autoregressive neural network modeling of time series

In this paper a hierarchical Bayesian learning scheme for autoregressive neural network models is shown, which overcomes the problem of identifying the separate linear and nonlinear parts in the network. We show how the identification can be carried out by defining suitable priors on the parameter space, which help the learning algorithms to avoid undesired parameter configurations. Some applications to synthetic data are shown to validate the proposed methodology.