Quo vadis, Bayesian identification?

The Bayesian identification of non-linear, non-Gaussian, non-stationary or non-parametric models is notoriously known as computer-intensive and not so lvable in a closed form. The paper outlines three major approaches to appro ximate Bayesian estimation, based on locally weighted smoothing of data, it erative and non-iterative Monte Carlo simulation and direct approximation o f an information 'distance' between the empirical and model distributions o f data. The information-based view of estimation is used throughout to give more insight into the methods and show their mutual relationship. Copyrigh t (C) 1999 John Wiley & Sons, Ltd.