Solidarity properties and a Doeblin decomposition for a class of non- Markovian stochastic processes (Q792698)

For a class of non-Markovian discrete parameter stochastic processes solidarity properties and a Doeblin decomposition are studied. Since any of those processes is associated with a certain general Markov chain whose transition probability function has a special form, the authors apply the theory of Markov chains with continuous components to this particular chain in order to get properties of the non-Markovian chain. The results are illustrated on a model closely related to learning theory.