Optimizing the asymptotic convergence rate of the Diaconis-Holmes-Neal sampler

Kranthi K. Gade, Michael L. Overton

Research output: Contribution to journalArticle

Abstract

We consider the problem of optimizing the asymptotic convergence rate of a parameter-dependent nonreversible Markov chain. We begin with a single-parameter case studied by Diaconis, Holmes and Neal and then introduce multiple parameters. We use nonsmooth analysis to investigate whether the presence of multiple parameters allows a faster asymptotic convergence rate, and argue that for a specific parameterization, it does not, at least locally.

Original languageEnglish (US)
Pages (from-to)382-403
Number of pages22
JournalAdvances in Applied Mathematics
Volume38
Issue number3
DOIs
StatePublished - Mar 2007

Fingerprint

Asymptotic Convergence
Parameterization
Markov processes
Convergence Rate
Nonsmooth Analysis
Markov chain
Dependent

Keywords

  • Diaconis-Holmes-Neal sampler
  • Markov chain optimization
  • Nonreversible Markov chains
  • Spectral functions
  • Variational analysis

ASJC Scopus subject areas

  • Applied Mathematics
  • Discrete Mathematics and Combinatorics

Cite this

Optimizing the asymptotic convergence rate of the Diaconis-Holmes-Neal sampler. / Gade, Kranthi K.; Overton, Michael L.

In: Advances in Applied Mathematics, Vol. 38, No. 3, 03.2007, p. 382-403.

Research output: Contribution to journalArticle

@article{2843712a123647e9b0cba74b298db1f9,
title = "Optimizing the asymptotic convergence rate of the Diaconis-Holmes-Neal sampler",
abstract = "We consider the problem of optimizing the asymptotic convergence rate of a parameter-dependent nonreversible Markov chain. We begin with a single-parameter case studied by Diaconis, Holmes and Neal and then introduce multiple parameters. We use nonsmooth analysis to investigate whether the presence of multiple parameters allows a faster asymptotic convergence rate, and argue that for a specific parameterization, it does not, at least locally.",
keywords = "Diaconis-Holmes-Neal sampler, Markov chain optimization, Nonreversible Markov chains, Spectral functions, Variational analysis",
author = "Gade, {Kranthi K.} and Overton, {Michael L.}",
year = "2007",
month = "3",
doi = "10.1016/j.aam.2006.05.003",
language = "English (US)",
volume = "38",
pages = "382--403",
journal = "Advances in Applied Mathematics",
issn = "0196-8858",
publisher = "Academic Press Inc.",
number = "3",

}

TY - JOUR

T1 - Optimizing the asymptotic convergence rate of the Diaconis-Holmes-Neal sampler

AU - Gade, Kranthi K.

AU - Overton, Michael L.

PY - 2007/3

Y1 - 2007/3

N2 - We consider the problem of optimizing the asymptotic convergence rate of a parameter-dependent nonreversible Markov chain. We begin with a single-parameter case studied by Diaconis, Holmes and Neal and then introduce multiple parameters. We use nonsmooth analysis to investigate whether the presence of multiple parameters allows a faster asymptotic convergence rate, and argue that for a specific parameterization, it does not, at least locally.

AB - We consider the problem of optimizing the asymptotic convergence rate of a parameter-dependent nonreversible Markov chain. We begin with a single-parameter case studied by Diaconis, Holmes and Neal and then introduce multiple parameters. We use nonsmooth analysis to investigate whether the presence of multiple parameters allows a faster asymptotic convergence rate, and argue that for a specific parameterization, it does not, at least locally.

KW - Diaconis-Holmes-Neal sampler

KW - Markov chain optimization

KW - Nonreversible Markov chains

KW - Spectral functions

KW - Variational analysis

UR - http://www.scopus.com/inward/record.url?scp=33846849268&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846849268&partnerID=8YFLogxK

U2 - 10.1016/j.aam.2006.05.003

DO - 10.1016/j.aam.2006.05.003

M3 - Article

AN - SCOPUS:33846849268

VL - 38

SP - 382

EP - 403

JO - Advances in Applied Mathematics

JF - Advances in Applied Mathematics

SN - 0196-8858

IS - 3

ER -