Bayesian estimation and prediction for inhomogeneous spatiotemporal log-gaussian cox processes using low-rank models, with application to criminal surveillance

Alexandre Rodrigues, Peter J. Diggle

Research output: Contribution to journalArticle

Abstract

In this article, we propose a method for conducting likelihood-based inference for a class of nonstationary spatiotemporal log-Gaussian Cox processes. The method uses convolution-based models to capture spatiotemporal correlation structure, is computationally feasible even for large datasets, and does not require knowledge of the underlying spatial intensity of the process.We describe an application to a surveillance system for detecting emergent spatiotemporal clusters of homicides in Belo Horizonte, Brazil, and discuss the advantages and drawbacks of our model-based approach by comparison with other spatiotemporal surveillance methods that have been proposed in the literature.

Original languageEnglish (US)
Pages (from-to)93-101
Number of pages9
JournalJournal of the American Statistical Association
Volume107
Issue number497
DOIs
StatePublished - 2012
Externally publishedYes

Fingerprint

Cox Process
Bayesian Prediction
Bayesian Estimation
Gaussian Process
Surveillance
Correlation Structure
Large Data Sets
Model
Convolution
Likelihood
Model-based
Cox process
Prediction
Bayesian estimation

Keywords

  • Convolution-based model
  • Likelihood-based inference
  • Spatiotemporal process
  • Surveillance system

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{837648b9c0a64818b26c06e4244b6c07,
title = "Bayesian estimation and prediction for inhomogeneous spatiotemporal log-gaussian cox processes using low-rank models, with application to criminal surveillance",
abstract = "In this article, we propose a method for conducting likelihood-based inference for a class of nonstationary spatiotemporal log-Gaussian Cox processes. The method uses convolution-based models to capture spatiotemporal correlation structure, is computationally feasible even for large datasets, and does not require knowledge of the underlying spatial intensity of the process.We describe an application to a surveillance system for detecting emergent spatiotemporal clusters of homicides in Belo Horizonte, Brazil, and discuss the advantages and drawbacks of our model-based approach by comparison with other spatiotemporal surveillance methods that have been proposed in the literature.",
keywords = "Convolution-based model, Likelihood-based inference, Spatiotemporal process, Surveillance system",
author = "Alexandre Rodrigues and Diggle, {Peter J.}",
year = "2012",
doi = "10.1080/01621459.2011.644496",
language = "English (US)",
volume = "107",
pages = "93--101",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "497",

}

TY - JOUR

T1 - Bayesian estimation and prediction for inhomogeneous spatiotemporal log-gaussian cox processes using low-rank models, with application to criminal surveillance

AU - Rodrigues, Alexandre

AU - Diggle, Peter J.

PY - 2012

Y1 - 2012

N2 - In this article, we propose a method for conducting likelihood-based inference for a class of nonstationary spatiotemporal log-Gaussian Cox processes. The method uses convolution-based models to capture spatiotemporal correlation structure, is computationally feasible even for large datasets, and does not require knowledge of the underlying spatial intensity of the process.We describe an application to a surveillance system for detecting emergent spatiotemporal clusters of homicides in Belo Horizonte, Brazil, and discuss the advantages and drawbacks of our model-based approach by comparison with other spatiotemporal surveillance methods that have been proposed in the literature.

AB - In this article, we propose a method for conducting likelihood-based inference for a class of nonstationary spatiotemporal log-Gaussian Cox processes. The method uses convolution-based models to capture spatiotemporal correlation structure, is computationally feasible even for large datasets, and does not require knowledge of the underlying spatial intensity of the process.We describe an application to a surveillance system for detecting emergent spatiotemporal clusters of homicides in Belo Horizonte, Brazil, and discuss the advantages and drawbacks of our model-based approach by comparison with other spatiotemporal surveillance methods that have been proposed in the literature.

KW - Convolution-based model

KW - Likelihood-based inference

KW - Spatiotemporal process

KW - Surveillance system

UR - http://www.scopus.com/inward/record.url?scp=84862838548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862838548&partnerID=8YFLogxK

U2 - 10.1080/01621459.2011.644496

DO - 10.1080/01621459.2011.644496

M3 - Article

VL - 107

SP - 93

EP - 101

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 497

ER -