Home > Research > Publications & Outputs > Restricted covariance priors with applications ...

Electronic data

Links

Text available via DOI:

View graph of relations

Restricted covariance priors with applications in spatial statistic

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Restricted covariance priors with applications in spatial statistic. / Smith, Theresa; Wakefield, Jon; Dobra, Adrian.
In: Bayesian Analysis, Vol. 10, No. 4, 12.2015, p. 965-990.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Smith, T, Wakefield, J & Dobra, A 2015, 'Restricted covariance priors with applications in spatial statistic', Bayesian Analysis, vol. 10, no. 4, pp. 965-990. https://doi.org/10.1214/14-BA927

APA

Smith, T., Wakefield, J., & Dobra, A. (2015). Restricted covariance priors with applications in spatial statistic. Bayesian Analysis, 10(4), 965-990. https://doi.org/10.1214/14-BA927

Vancouver

Smith T, Wakefield J, Dobra A. Restricted covariance priors with applications in spatial statistic. Bayesian Analysis. 2015 Dec;10(4):965-990. Epub 2015 Feb 4. doi: 10.1214/14-BA927

Author

Smith, Theresa ; Wakefield, Jon ; Dobra, Adrian. / Restricted covariance priors with applications in spatial statistic. In: Bayesian Analysis. 2015 ; Vol. 10, No. 4. pp. 965-990.

Bibtex

@article{5ea852b719b441c9b429560c718e2bff,
title = "Restricted covariance priors with applications in spatial statistic",
abstract = "We present a Bayesian model for area-level count data that uses Gaussian random effects with a novel type of G-Wishart prior on the inverse variance–covariance matrix. Specifically, we introduce a new distribution called the truncated G-Wishart distribution that has support over precision matrices that lead to positive associations between the random effects of neighboring regions while preserving conditional independence of non-neighboring regions. We describe Markov chain Monte Carlo sampling algorithms for the truncated G-Wishart prior in a disease mapping context and compare our results to Bayesian hierarchical models based on intrinsic autoregression priors. A simulation study illustrates that using the truncated G-Wishart prior improves over the intrinsic autoregressive priors when there are discontinuities in the disease risk surface. The new model is applied to an analysis of cancer incidence data in Washington State.",
author = "Theresa Smith and Jon Wakefield and Adrian Dobra",
year = "2015",
month = dec,
doi = "10.1214/14-BA927",
language = "English",
volume = "10",
pages = "965--990",
journal = "Bayesian Analysis",
issn = "1936-0975",
publisher = "Carnegie Mellon University",
number = "4",

}

RIS

TY - JOUR

T1 - Restricted covariance priors with applications in spatial statistic

AU - Smith, Theresa

AU - Wakefield, Jon

AU - Dobra, Adrian

PY - 2015/12

Y1 - 2015/12

N2 - We present a Bayesian model for area-level count data that uses Gaussian random effects with a novel type of G-Wishart prior on the inverse variance–covariance matrix. Specifically, we introduce a new distribution called the truncated G-Wishart distribution that has support over precision matrices that lead to positive associations between the random effects of neighboring regions while preserving conditional independence of non-neighboring regions. We describe Markov chain Monte Carlo sampling algorithms for the truncated G-Wishart prior in a disease mapping context and compare our results to Bayesian hierarchical models based on intrinsic autoregression priors. A simulation study illustrates that using the truncated G-Wishart prior improves over the intrinsic autoregressive priors when there are discontinuities in the disease risk surface. The new model is applied to an analysis of cancer incidence data in Washington State.

AB - We present a Bayesian model for area-level count data that uses Gaussian random effects with a novel type of G-Wishart prior on the inverse variance–covariance matrix. Specifically, we introduce a new distribution called the truncated G-Wishart distribution that has support over precision matrices that lead to positive associations between the random effects of neighboring regions while preserving conditional independence of non-neighboring regions. We describe Markov chain Monte Carlo sampling algorithms for the truncated G-Wishart prior in a disease mapping context and compare our results to Bayesian hierarchical models based on intrinsic autoregression priors. A simulation study illustrates that using the truncated G-Wishart prior improves over the intrinsic autoregressive priors when there are discontinuities in the disease risk surface. The new model is applied to an analysis of cancer incidence data in Washington State.

U2 - 10.1214/14-BA927

DO - 10.1214/14-BA927

M3 - Journal article

VL - 10

SP - 965

EP - 990

JO - Bayesian Analysis

JF - Bayesian Analysis

SN - 1936-0975

IS - 4

ER -