Home > Research > Publications & Outputs > Density Estimation for the Metropolis–Hastings ...
View graph of relations

Density Estimation for the Metropolis–Hastings Algorithm.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Density Estimation for the Metropolis–Hastings Algorithm. / Sköld, M.; Roberts, G. O.
In: Scandinavian Journal of Statistics, Vol. 30, No. 4, 12.2003, p. 699-718.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Sköld, M & Roberts, GO 2003, 'Density Estimation for the Metropolis–Hastings Algorithm.', Scandinavian Journal of Statistics, vol. 30, no. 4, pp. 699-718. https://doi.org/10.1111/1467-9469.00359

APA

Sköld, M., & Roberts, G. O. (2003). Density Estimation for the Metropolis–Hastings Algorithm. Scandinavian Journal of Statistics, 30(4), 699-718. https://doi.org/10.1111/1467-9469.00359

Vancouver

Sköld M, Roberts GO. Density Estimation for the Metropolis–Hastings Algorithm. Scandinavian Journal of Statistics. 2003 Dec;30(4):699-718. doi: 10.1111/1467-9469.00359

Author

Sköld, M. ; Roberts, G. O. / Density Estimation for the Metropolis–Hastings Algorithm. In: Scandinavian Journal of Statistics. 2003 ; Vol. 30, No. 4. pp. 699-718.

Bibtex

@article{5c23fa4e2e524becb95300a355ebdbca,
title = "Density Estimation for the Metropolis–Hastings Algorithm.",
abstract = "Kernel density estimation is an important tool in visualizing posterior densities from Markov chain Monte Carlo output. It is well known that when smooth transition densities exist, the asymptotic properties of the estimator agree with those for independent data. In this paper, we show that because of the rejection step of the Metropolis–Hastings algorithm, this is no longer true and the asymptotic variance will depend on the probability of accepting a proposed move. We find an expression for this variance and apply the result to algorithms for automatic bandwidth selection.",
author = "M. Sk{\"o}ld and Roberts, {G. O.}",
year = "2003",
month = dec,
doi = "10.1111/1467-9469.00359",
language = "English",
volume = "30",
pages = "699--718",
journal = "Scandinavian Journal of Statistics",
issn = "1467-9469",
publisher = "Blackwell-Wiley",
number = "4",

}

RIS

TY - JOUR

T1 - Density Estimation for the Metropolis–Hastings Algorithm.

AU - Sköld, M.

AU - Roberts, G. O.

PY - 2003/12

Y1 - 2003/12

N2 - Kernel density estimation is an important tool in visualizing posterior densities from Markov chain Monte Carlo output. It is well known that when smooth transition densities exist, the asymptotic properties of the estimator agree with those for independent data. In this paper, we show that because of the rejection step of the Metropolis–Hastings algorithm, this is no longer true and the asymptotic variance will depend on the probability of accepting a proposed move. We find an expression for this variance and apply the result to algorithms for automatic bandwidth selection.

AB - Kernel density estimation is an important tool in visualizing posterior densities from Markov chain Monte Carlo output. It is well known that when smooth transition densities exist, the asymptotic properties of the estimator agree with those for independent data. In this paper, we show that because of the rejection step of the Metropolis–Hastings algorithm, this is no longer true and the asymptotic variance will depend on the probability of accepting a proposed move. We find an expression for this variance and apply the result to algorithms for automatic bandwidth selection.

U2 - 10.1111/1467-9469.00359

DO - 10.1111/1467-9469.00359

M3 - Journal article

VL - 30

SP - 699

EP - 718

JO - Scandinavian Journal of Statistics

JF - Scandinavian Journal of Statistics

SN - 1467-9469

IS - 4

ER -