Markov-switching state-space models with applications to neuroimaging

State-space models (SSM) with Markov switching offer a powerful framework for detecting multiple regimes in time series, analyzing mutual dependence and dynamics within regimes, assessing transitions between regimes. These however present considerable computational challenges due to the exponential number of possible regime sequences account for. In addition, high dimensionality series can hinder likelihood-based inference. To address these challenges, novel statistical methods Markov-switching SSMs are proposed using maximum likelihood estimation, Expectation-Maximization (EM), parametric bootstrap. Solutions developed initializing EM algorithm, accelerating convergence, conducting methods, which ideally suited massive spatio-temporal data such as brain signals, evaluated simulations applications EEG studies epilepsy motor imagery presented. • Novel estimation bootstrap state-space models. Highly optimized software enables flexible model specification accommodates long and/or high-dimensional series. Applications brain-computer interfaces deliver meaningful scientific results.