Bayesian nonstationary autoregressive models for biomedical signal analysis
Michael J. Cassidy, W.D. Penny · IEEE Transactions on Biomedical Engineering · 2002
We describe a variational Bayesian algorithm for the estimation of a multivariate autoregressive model with time-varying coefficients that adapt according to a linear dynamical system. The algorithm allows for time and frequency domain characterization of nonstationary multivariate signals and is especially suited to the analysis of event-related data. Results are presented on synthetic data and real electroencephalogram data recorded in event-related desynchronization and photic synchronization scenarios.