Hidden Markov Models in Time Series, with Applications in Economics

Hidden Markov models are mixture models with sequential dependence or persistence in the mixture distribution. For a finite number G of components, persistence in distribution is induced by specifying a latent component indicator which follows a Markov process. The 310transition probabilities for the Markov process may either be time-invariant or time-varying. In the latter case, hidden Markov models extend mixture of experts models (see Chapter 12) by introducing persistence in the mixtures.