A Markov chain model is one appropriate probability model for time series data when the observation at any point in time is the state or category into which the unit being observed falls. Although the Markov probability model is of relatively recent vintage, it has been used as a basis for describing the generation of sets of data over a wide variety of subject matter areas.