Sampling-reconstruction of markov processes with a finite number of states

Statistical procedure for sampling-reconstruction of Markov processes with a finite number of states is developed. Special features of the sampling-reconstruction procedures for such processes compared to continuous processes are discussed. The problem of determining the sampling interval is formulated and solved under the following two constraints: on the variance of the jump moment estimate and on the probability of missing a transition. An example illustrates the application of the described methodology.