Markov chain monte carlo estimation of quantiles

Charles R. Doss; James M. Flegal; Galin L. Jones; Ronald C. Neath

doi:10.1214/14-EJS957

Markov chain monte carlo estimation of quantiles

Charles R. Doss, James M. Flegal, Galin L. Jones, Ronald C. Neath

Statistics (Twin Cities)

Research output: Contribution to journal › Article › peer-review

23 Scopus citations

Abstract

We consider quantile estimation using Markov chain Monte Carlo and establish conditions under which the sampling distribution of the Monte Carlo error is approximately Normal. Further, we investigate techniques to estimate the associated asymptotic variance, which enables construction of an asymptotically valid interval estimator. Finally, we ex- plore the finite sample properties of these methods through examples and provide some recommendations to practitioners.

Original language	English (US)
Pages (from-to)	2448-2478
Number of pages	31
Journal	Electronic Journal of Statistics
Volume	8
DOIs	https://doi.org/10.1214/14-EJS957
State	Published - 2015

Bibliographical note

Publisher Copyright:
© 2014, Institute of Mathematical Statistics. All rights reserved.

Keywords

Batch means
Central limit theorem
Markov chain
Monte carlo
Quantile estimation
Regeneration

Publisher link

10.1214/14-EJS957

Find It

Find It at U of M Libraries

Cite this

@article{45d56f3d3f394d84b60f6f3bbccff676,

title = "Markov chain monte carlo estimation of quantiles",

abstract = "We consider quantile estimation using Markov chain Monte Carlo and establish conditions under which the sampling distribution of the Monte Carlo error is approximately Normal. Further, we investigate techniques to estimate the associated asymptotic variance, which enables construction of an asymptotically valid interval estimator. Finally, we ex- plore the finite sample properties of these methods through examples and provide some recommendations to practitioners.",

keywords = "Batch means, Central limit theorem, Markov chain, Monte carlo, Quantile estimation, Regeneration",

author = "Doss, {Charles R.} and Flegal, {James M.} and Jones, {Galin L.} and Neath, {Ronald C.}",

year = "2015",

doi = "10.1214/14-EJS957",

language = "English (US)",

volume = "8",

pages = "2448--2478",

journal = "Electronic Journal of Statistics",

issn = "1935-7524",

publisher = "Institute of Mathematical Statistics",

}

TY - JOUR

T1 - Markov chain monte carlo estimation of quantiles

AU - Doss, Charles R.

AU - Flegal, James M.

AU - Jones, Galin L.

AU - Neath, Ronald C.

PY - 2015

Y1 - 2015

N2 - We consider quantile estimation using Markov chain Monte Carlo and establish conditions under which the sampling distribution of the Monte Carlo error is approximately Normal. Further, we investigate techniques to estimate the associated asymptotic variance, which enables construction of an asymptotically valid interval estimator. Finally, we ex- plore the finite sample properties of these methods through examples and provide some recommendations to practitioners.

AB - We consider quantile estimation using Markov chain Monte Carlo and establish conditions under which the sampling distribution of the Monte Carlo error is approximately Normal. Further, we investigate techniques to estimate the associated asymptotic variance, which enables construction of an asymptotically valid interval estimator. Finally, we ex- plore the finite sample properties of these methods through examples and provide some recommendations to practitioners.

KW - Batch means

KW - Central limit theorem

KW - Markov chain

KW - Monte carlo

KW - Quantile estimation

KW - Regeneration

UR - http://www.scopus.com/inward/record.url?scp=84934930312&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84934930312&partnerID=8YFLogxK

U2 - 10.1214/14-EJS957

DO - 10.1214/14-EJS957

M3 - Article

AN - SCOPUS:84934930312

SN - 1935-7524

VL - 8

SP - 2448

EP - 2478

JO - Electronic Journal of Statistics

JF - Electronic Journal of Statistics

ER -

Markov chain monte carlo estimation of quantiles

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this