State Space Markov Switching Models Using Wavelets

We propose a state space model with Markov switching, whose regimes are associated with the model parameters and regime transition probabilities are time-dependent. The estimation is based on maximum likelihood method using the EM algorithm. The distribution of the estimators is assessed using bootstrap. To evaluate the state variables and regime probabilities, the Kalman filter and a probability filter procedure conditional to each possible regime at each instant are used. This procedure is illustrated with simulated data and the United States monthly industrial production index from January 1960 to January 1995.