Seven things to remember about hidden Markov models: A tutorial on Markovian models for time series

This paper provides a tutorial on key issues in hidden Markov modeling. Hidden Markov models have become very popular models for time series and longitudinal data in recent years due to a combination of (relative) simplicity and flexibility in adapting the model to novel situations. The tutorial covers the conceptual description of the model, estimation of parameters through maximum likelihood, and ends with an application to real data illustrating the possibilities. (C) 2011 Elsevier Inc. All rights reserved.