Abstract:
|
Hidden semi-Markov models (HSMMs) extend the hidden Markov model (HMM) by explicitly modeling the time spent in each state. In a discrete-time HSMM, the duration in each state can be modeled with a zero-truncated Poisson distribution, where the intensity may be state-specific but constant in time. We extend the HSMM by allowing the state-specific intensities to vary in time and model them as a function of known covariates. In particular, the intensity at each state transition is a function of covariates observed over a period of time leading up to the transition. Model inference is obtained in Bayesian framework and our HSMM with time-varying intensities can be applied broadly in applications ranging from environmental, to speech recognition, to sports.
|