We develop an algorithm for simulating “perfect”
random samples from the invariant measure of a Harris recurrent Markov
chain.The method uses backward coupling of embedded regeneration times and
works most effectively for stochastically monotone chains, where paths may be
sandwiched between “upper” and “lower” processes.
We give an approach to finding analytic bounds on the backward coupling times
in the stochastically monotone case. An application to storage models is
given.
Publié le : 2001-05-14
Classification:
Irreducible Markov chains,
invariant measures,
geometric ergodicity,
backward coupling,
coupling from the past,
exact sampling,
perfect sampling,
queues,
storage models,
60J10,
60K05,
60K30
@article{1015345299,
author = {Corcoran, J. N. and Tweedie, R. L.},
title = {Perfect sampling of ergodic Harris chains},
journal = {Ann. Appl. Probab.},
volume = {11},
number = {2},
year = {2001},
pages = { 438-451},
language = {en},
url = {http://dml.mathdoc.fr/item/1015345299}
}
Corcoran, J. N.; Tweedie, R. L. Perfect sampling of ergodic Harris chains. Ann. Appl. Probab., Tome 11 (2001) no. 2, pp. 438-451. http://gdmltest.u-ga.fr/item/1015345299/