Segmentation of switching dynamics with a Hidden Markov Model of neural prediction experts

We discuss a framework for modeling the switching dynamics of a time series based on hidden Markov models (HMM) of prediction experts, here neural networks. Learning is treated as a maximum likelihood problem. In particular, we present an Expectation-Maximization (EM) algorithm for adjusting the expert parameters as well as the HMM transition probabilities. Based on this algorithm, we develop a heuristic that achieves a hard segmentation of the time series into distinct dynamical modes and the simultaneous specialization of the prediction experts on the segments. We present examples of the application of this algorithm to the segmentation of arti cial and nancial time series.