
Using a Markov Chain to Construct a Tractable Approximation of an Intractable Probability Distribution
Hobert, James P.; Jones, Galin L.; Robert, Christian P. (2006), Using a Markov Chain to Construct a Tractable Approximation of an Intractable Probability Distribution, Scandinavian Journal of Statistics, 33, 1, p. 37-51. http://dx.doi.org/10.1111/j.1467-9469.2006.00467.x
Voir/Ouvrir
Type
Article accepté pour publication ou publiéDate
2006Nom de la revue
Scandinavian Journal of StatisticsVolume
33Numéro
1Éditeur
Wiley
Pages
37-51
Identifiant publication
Métadonnées
Afficher la notice complèteRésumé (EN)
Let π denote an intractable probability distribution that we would like to explore. Suppose that we have a positive recurrent, irreducible Markov chain that satisfies a minorization condition and has π as its invariant measure. We provide a method of using simulations from the Markov chain to construct a statistical estimate of π from which it is straightforward to sample. We show that this estimate is ‘strongly consistent’ in the sense that the total variation distance between the estimate and π converges to 0 almost surely as the number of simulations grows. Moreover, we use some recently developed asymptotic results to provide guidance as to how much simulation is necessary. Draws from the estimate can be used to approximate features of π or as intelligent starting values for the original Markov chain. We illustrate our methods with two examples.Mots-clés
burn-in; Gibbs sampler; minorization condition; mixture representation; Monte Carlo; regeneration; split chainPublications associées
Affichage des éléments liés par titre et auteur.
-
Hobert, James P.; Robert, Christian P. (2004) Article accepté pour publication ou publié
-
Robert, Christian P.; Roy, Vivekananda; Hobert, James P. (2011) Article accepté pour publication ou publié
-
Robert, Christian P.; Hobert, James P. (2001) Document de travail / Working paper
-
Casella, George; Robert, Christian P. (2011) Article accepté pour publication ou publié
-
Robert, Christian P.; Roberts, Gareth (2021) Article accepté pour publication ou publié