Markov chain transition probabilities and experimental data

Markov theory is used in many applications to provide a probabilistic framework for modeling dynamic behavior. This theory is attractive because (1) it is the probabilistic analogue to the classical physics approach to dynamics, (2) it is easily recast into a computational form, and (3) the state transition matrix is an ideal probabilistic counterpart to the deterministic transfer matrix concept. The determination of the transition matrix is the heart of the method. An approach by which experimental data can be used to estimate the elements of this probability matrix is developed.