The role of simulation in modelling spatially correlated data

N. H. Augustin, M. A. Mugglestone, S. T. Buckland

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

Three approaches to modelling spatial data in which simulation plays a vital role are described and illustrated with examples. The first approach uses flexible regression models, such as generalized additive models, together with locational covariates to fit a surface to spatial data. We show how the bootstrap can be used to quantify the effects of model selection uncertainty and to avoid oversmoothing. The second approach, which is appropriate for binary data, allows for local spatial correlation by the inclusion in a logistic regression model of a covariate derived from neighbouring values of the response variable. The resulting autologistic model can be fitted to survey data obtained from a random sample of sites by incorporating the Gibbs sampler into the modelling procedure. We show this modelling strategy can be used not only to fit the autologistic model to sites included in the survey, but also to estimate the probability that a certain species is present in the unsurveyed sites. Our third approach relates to the analysis of spatio-temporal data. Here we model the distribution of a plant of animal species as a function of the distribution at an earlier time point. The bootstrap is used to estimate parameters and quantify their precision.

Original languageEnglish
Pages (from-to)175-196
Number of pages22
JournalEnvironmetrics
Volume9
Issue number2
DOIs
Publication statusPublished - 1 Mar 1998

Keywords

  • Autologistic model
  • Bootstrap
  • Generalized additive model
  • Generalized linear model
  • Gibbs sampler
  • Model uncertainty
  • Red deer
  • Simulated inference
  • Spatial distribution
  • Spatio-temporal model

ASJC Scopus subject areas

  • Statistics and Probability
  • Ecological Modelling

Cite this

The role of simulation in modelling spatially correlated data. / Augustin, N. H.; Mugglestone, M. A.; Buckland, S. T.

In: Environmetrics, Vol. 9, No. 2, 01.03.1998, p. 175-196.

Research output: Contribution to journalArticle

Augustin, N. H. ; Mugglestone, M. A. ; Buckland, S. T. / The role of simulation in modelling spatially correlated data. In: Environmetrics. 1998 ; Vol. 9, No. 2. pp. 175-196.
@article{449db9408c874a4f9568699a4c5b28ad,
title = "The role of simulation in modelling spatially correlated data",
abstract = "Three approaches to modelling spatial data in which simulation plays a vital role are described and illustrated with examples. The first approach uses flexible regression models, such as generalized additive models, together with locational covariates to fit a surface to spatial data. We show how the bootstrap can be used to quantify the effects of model selection uncertainty and to avoid oversmoothing. The second approach, which is appropriate for binary data, allows for local spatial correlation by the inclusion in a logistic regression model of a covariate derived from neighbouring values of the response variable. The resulting autologistic model can be fitted to survey data obtained from a random sample of sites by incorporating the Gibbs sampler into the modelling procedure. We show this modelling strategy can be used not only to fit the autologistic model to sites included in the survey, but also to estimate the probability that a certain species is present in the unsurveyed sites. Our third approach relates to the analysis of spatio-temporal data. Here we model the distribution of a plant of animal species as a function of the distribution at an earlier time point. The bootstrap is used to estimate parameters and quantify their precision.",
keywords = "Autologistic model, Bootstrap, Generalized additive model, Generalized linear model, Gibbs sampler, Model uncertainty, Red deer, Simulated inference, Spatial distribution, Spatio-temporal model",
author = "Augustin, {N. H.} and Mugglestone, {M. A.} and Buckland, {S. T.}",
year = "1998",
month = "3",
day = "1",
doi = "10.1002/(SICI)1099-095X(199803/04)9:2<175::AID-ENV294>3.0.CO;2-2",
language = "English",
volume = "9",
pages = "175--196",
journal = "Environmetrics",
issn = "1180-4009",
publisher = "John Wiley and Sons Inc.",
number = "2",

}

TY - JOUR

T1 - The role of simulation in modelling spatially correlated data

AU - Augustin, N. H.

AU - Mugglestone, M. A.

AU - Buckland, S. T.

PY - 1998/3/1

Y1 - 1998/3/1

N2 - Three approaches to modelling spatial data in which simulation plays a vital role are described and illustrated with examples. The first approach uses flexible regression models, such as generalized additive models, together with locational covariates to fit a surface to spatial data. We show how the bootstrap can be used to quantify the effects of model selection uncertainty and to avoid oversmoothing. The second approach, which is appropriate for binary data, allows for local spatial correlation by the inclusion in a logistic regression model of a covariate derived from neighbouring values of the response variable. The resulting autologistic model can be fitted to survey data obtained from a random sample of sites by incorporating the Gibbs sampler into the modelling procedure. We show this modelling strategy can be used not only to fit the autologistic model to sites included in the survey, but also to estimate the probability that a certain species is present in the unsurveyed sites. Our third approach relates to the analysis of spatio-temporal data. Here we model the distribution of a plant of animal species as a function of the distribution at an earlier time point. The bootstrap is used to estimate parameters and quantify their precision.

AB - Three approaches to modelling spatial data in which simulation plays a vital role are described and illustrated with examples. The first approach uses flexible regression models, such as generalized additive models, together with locational covariates to fit a surface to spatial data. We show how the bootstrap can be used to quantify the effects of model selection uncertainty and to avoid oversmoothing. The second approach, which is appropriate for binary data, allows for local spatial correlation by the inclusion in a logistic regression model of a covariate derived from neighbouring values of the response variable. The resulting autologistic model can be fitted to survey data obtained from a random sample of sites by incorporating the Gibbs sampler into the modelling procedure. We show this modelling strategy can be used not only to fit the autologistic model to sites included in the survey, but also to estimate the probability that a certain species is present in the unsurveyed sites. Our third approach relates to the analysis of spatio-temporal data. Here we model the distribution of a plant of animal species as a function of the distribution at an earlier time point. The bootstrap is used to estimate parameters and quantify their precision.

KW - Autologistic model

KW - Bootstrap

KW - Generalized additive model

KW - Generalized linear model

KW - Gibbs sampler

KW - Model uncertainty

KW - Red deer

KW - Simulated inference

KW - Spatial distribution

KW - Spatio-temporal model

UR - http://www.scopus.com/inward/record.url?scp=0031956948&partnerID=8YFLogxK

U2 - 10.1002/(SICI)1099-095X(199803/04)9:2<175::AID-ENV294>3.0.CO;2-2

DO - 10.1002/(SICI)1099-095X(199803/04)9:2<175::AID-ENV294>3.0.CO;2-2

M3 - Article

VL - 9

SP - 175

EP - 196

JO - Environmetrics

JF - Environmetrics

SN - 1180-4009

IS - 2

ER -