Premium
Markov Sampling
Author(s) -
Deshmukh S.R.
Publication year - 2000
Publication title -
australian and new zealand journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 1369-1473
DOI - 10.1111/1467-842x.00130
Subject(s) - markov chain , mathematics , markov renewal process , markov process , markov property , variable order markov model , markov kernel , identifiability , markov chain mixing time , continuous time markov chain , markov model , additive markov chain , sampling (signal processing) , examples of markov chains , time reversibility , finite state , statistics , computer science , filter (signal processing) , computer vision
A discrete parameter stochastic process is observed at epochs of visits to a specified state in an independent two‐state Markov chain. It is established that the family of finite dimensional distributions of the process derived in this way, referred to as Markov sampling, uniquely determines the stochastic structure of the original process. Using this identifiability, it is shown that if the derived process is Markov, then the original process is also Markov and if the derived process is strictly stationary then so is the original.