Careful prior specification avoids incautious inference for log-Gaussian Cox point processes

Sigrunn H. Sørbye (Corresponding Author), Janine B. Illian, Daniel P. Simpson, David F. R. P. Burslem, Håvard Rue

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Hyperprior specifications for random fields in spatial point process modelling
can have a major impact on the results. In fitting log-Gaussian Cox processes to rainforest tree species, we consider a reparameterised model combining a spatially structured and an unstructured random field into a single component. This component has one hyperparameter accounting for marginal variance, while an additional hyperparameter governs the fraction of the variance explained by the spatially structured effect. This facilitates interpretation of the hyperparameters and significance of covariates is studied for a range of hyperprior specifications. Appropriate scaling makes the analysis invariant to grid resolution.
Original languageEnglish
Pages (from-to)543-564
Number of pages20
JournalJournal of the royal statistical society series c-Applied statistics
Volume68
Issue number3
Early online date2 Nov 2018
DOIs
Publication statusPublished - Apr 2019

Fingerprint

Cox Process
Hyperparameters
Point Process
Specification
Random Field
Spatial Point Process
Gaussian Process
Covariates
Scaling
Grid
Invariant
Range of data
Inference
Random field
Point process
Model
Cox process
Rain forest

Keywords

  • Bayesian analysis
  • spatial point process
  • penalised complexity prior
  • R-INLA
  • Spatial modelling
  • Penalized complexity prior
  • Spatial point process

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Careful prior specification avoids incautious inference for log-Gaussian Cox point processes. / Sørbye, Sigrunn H. (Corresponding Author); Illian, Janine B.; Simpson, Daniel P.; Burslem, David F. R. P.; Rue, Håvard.

In: Journal of the royal statistical society series c-Applied statistics, Vol. 68, No. 3, 04.2019, p. 543-564.

Research output: Contribution to journalArticle

@article{012f366d9ce14758830aba5483147696,
title = "Careful prior specification avoids incautious inference for log-Gaussian Cox point processes",
abstract = "Hyperprior specifications for random fields in spatial point process modellingcan have a major impact on the results. In fitting log-Gaussian Cox processes to rainforest tree species, we consider a reparameterised model combining a spatially structured and an unstructured random field into a single component. This component has one hyperparameter accounting for marginal variance, while an additional hyperparameter governs the fraction of the variance explained by the spatially structured effect. This facilitates interpretation of the hyperparameters and significance of covariates is studied for a range of hyperprior specifications. Appropriate scaling makes the analysis invariant to grid resolution.",
keywords = "Bayesian analysis, spatial point process, penalised complexity prior, R-INLA, Spatial modelling, Penalized complexity prior, Spatial point process",
author = "S{\o}rbye, {Sigrunn H.} and Illian, {Janine B.} and Simpson, {Daniel P.} and Burslem, {David F. R. P.} and H{\aa}vard Rue",
note = "The BCI forest dynamics research project was founded by S.P. Hubbell and R.B. Foster and is now managed by R. Condit, S. Lao, and R. Perez under the Center for Tropical Forest Science and the Smithsonian Tropical Research in Panama. Numerous organizations have provided funding, principally the U.S. National Science Foundation, and hundreds of field workers have contributed. The data used can be requested and generally granted at http://ctfs.si.edudatarequest. Kriged estimates for concentration of the soil nutrients were downloaded from http://ctfs.si.edu/webatlas/datasets/bci/soilmaps/BCIsoil.html. We acknowledge the principal investigators that were responsible for collecting and analysing the soil maps (Jim Dallin, Robert John, Kyle Harms, Robert Stallard and Joe Yavitt), the funding sources (NSF DEB021104,021115, 0212284,0212818 and OISE 0314581, STRI Soils Initiative and CTFS) and field assistants (Paolo Segre and Juan Di Trani).",
year = "2019",
month = "4",
doi = "10.1111/rssc.12321",
language = "English",
volume = "68",
pages = "543--564",
journal = "Journal of the royal statistical society series c-Applied statistics",
issn = "0035-9254",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Careful prior specification avoids incautious inference for log-Gaussian Cox point processes

AU - Sørbye, Sigrunn H.

AU - Illian, Janine B.

AU - Simpson, Daniel P.

AU - Burslem, David F. R. P.

AU - Rue, Håvard

N1 - The BCI forest dynamics research project was founded by S.P. Hubbell and R.B. Foster and is now managed by R. Condit, S. Lao, and R. Perez under the Center for Tropical Forest Science and the Smithsonian Tropical Research in Panama. Numerous organizations have provided funding, principally the U.S. National Science Foundation, and hundreds of field workers have contributed. The data used can be requested and generally granted at http://ctfs.si.edudatarequest. Kriged estimates for concentration of the soil nutrients were downloaded from http://ctfs.si.edu/webatlas/datasets/bci/soilmaps/BCIsoil.html. We acknowledge the principal investigators that were responsible for collecting and analysing the soil maps (Jim Dallin, Robert John, Kyle Harms, Robert Stallard and Joe Yavitt), the funding sources (NSF DEB021104,021115, 0212284,0212818 and OISE 0314581, STRI Soils Initiative and CTFS) and field assistants (Paolo Segre and Juan Di Trani).

PY - 2019/4

Y1 - 2019/4

N2 - Hyperprior specifications for random fields in spatial point process modellingcan have a major impact on the results. In fitting log-Gaussian Cox processes to rainforest tree species, we consider a reparameterised model combining a spatially structured and an unstructured random field into a single component. This component has one hyperparameter accounting for marginal variance, while an additional hyperparameter governs the fraction of the variance explained by the spatially structured effect. This facilitates interpretation of the hyperparameters and significance of covariates is studied for a range of hyperprior specifications. Appropriate scaling makes the analysis invariant to grid resolution.

AB - Hyperprior specifications for random fields in spatial point process modellingcan have a major impact on the results. In fitting log-Gaussian Cox processes to rainforest tree species, we consider a reparameterised model combining a spatially structured and an unstructured random field into a single component. This component has one hyperparameter accounting for marginal variance, while an additional hyperparameter governs the fraction of the variance explained by the spatially structured effect. This facilitates interpretation of the hyperparameters and significance of covariates is studied for a range of hyperprior specifications. Appropriate scaling makes the analysis invariant to grid resolution.

KW - Bayesian analysis

KW - spatial point process

KW - penalised complexity prior

KW - R-INLA

KW - Spatial modelling

KW - Penalized complexity prior

KW - Spatial point process

UR - http://www.scopus.com/inward/record.url?scp=85055945954&partnerID=8YFLogxK

U2 - 10.1111/rssc.12321

DO - 10.1111/rssc.12321

M3 - Article

VL - 68

SP - 543

EP - 564

JO - Journal of the royal statistical society series c-Applied statistics

JF - Journal of the royal statistical society series c-Applied statistics

SN - 0035-9254

IS - 3

ER -