A TIME SERIES MODEL WITH SUDDENLY CHANGING PARAMETERS
Jiří Anděl · Journal of Time Series Analysis · 1993
Abstract. A process Xt=θt+e, is investigated where et is a strict white noise and θt is a Markov chain with two real states. A realization of Xt fluctuates around two levels which correspond to the two states of θt. Formulae for the extrapolation of the process Xt and for the mean square error of the extrapolation are derived. The moments of Xt and its covariance function are calculated. The results are used to derive moment estimators for the parameters of the model.