Skip to main content

An extension of Peskun and Tierney Orderings to continuous time Markov chains

Leisen, Fabrizio, Mira, Antonietta (2008) An extension of Peskun and Tierney Orderings to continuous time Markov chains. Statistica Sinica, 18 (4). pp. 1641-1651. ISSN 1017-0405. (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided)

The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided. (Contact us about this Publication)
Official URL
http://www3.stat.sinica.edu.tw/statistica/J18N4/J1...

Abstract

Peskun ordering is a partial ordering defined on the space of transition matrices of discrete time Markov chains. If the Markov chains are reversible with respect to a common stationary distribution $\pi$, Peskun ordering implies an ordering on the asymptotic variances of the resulting Markov chain Monte Carlo estimators of integrals with respect to $\pi$. Peskun ordering is also relevant in the framework of time-invariant estimating equations in that it provides a necessary condition for ordering the asymptotic variances of the resulting estimators. Tierney ordering extends Peskun ordering from finite to general state spaces. In this paper Peskun and Tierney orderings are extended from discrete time to continuous time Markov chains.

Item Type: Article
Uncontrolled keywords: Asymptotic variance, covariance ordering, efficiency ordering, MCMC, time-invariance estimating equations.
Subjects: H Social Sciences > HA Statistics
Q Science > QA Mathematics (inc Computing science) > QA273 Probabilities
Divisions: Faculties > Sciences > School of Mathematics Statistics and Actuarial Science
Faculties > Sciences > School of Mathematics Statistics and Actuarial Science > Statistics
Depositing User: Fabrizio Leisen
Date Deposited: 24 Dec 2013 13:51 UTC
Last Modified: 01 Aug 2019 10:36 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/37693 (The current URI for this page, for reference purposes)
Leisen, Fabrizio: https://orcid.org/0000-0002-2460-6176
  • Depositors only (login required):