Open AccessHigh-order Vector Markov Chain with Partial Connections in Data Analysis
Author(s) -
Yu. S. Kharin,
M. V. Maltsew
Publication year - 2017
Publication title -
österreichische zeitschrift für statistik
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.342
H-Index - 9
ISSN - 1026-597X
DOI - 10.17713/ajs.v46i3-4.669
Subject(s) - markov chain , ergodicity , mathematics , discrete phase type distribution , continuous time markov chain , additive markov chain , markov property , probabilistic logic , variable order markov model , markov model , probability distribution , estimator , conditional probability distribution , balance equation , markov chain mixing time , statistical physics , statistics , physics
A new mathematical model for discrete time series is proposed: homogenous vector Markov chain of the order s with partial connections. Conditional probability distribution for this model is determined only by a few components of previous vector states. Probabilistic properties of the model are given: ergodicity conditions and conditions under which the stationary probability distribution is uniform. Consistent statistical estimators for model parameters are constructed.