A multinomial logistic mixed model for the prediction of categorical spatial data

Guofeng Cao, Phaedon C. Kyriakidis, Michael Goodchild

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

In this article, the prediction problem of categorical spatial data, that is, the estimation of class occurrence probability for (target) locations with unknown class labels given observed class labels at sample (source) locations, is analyzed in the framework of generalized linear mixed models, where intermediate, latent (unobservable) spatially correlated Gaussian variables (random effects) are assumed for the observable non-Gaussian responses to account for spatial dependence information. Within such a framework, a spatial multinomial logistic mixed model is proposed specifically to model categorical spatial data. Analogous to the dual form of kriging family, the proposed model is represented as a multinomial logistic function of spatial covariances between target and source locations. The associated inference problems, such as estimation of parameters and choice of the spatial covariance function for latent variables, and the connection of the proposed model with other methods, such as the indicator variants of the kriging family (indicator kriging and indicator cokriging) and Bayesian maximum entropy, are discussed in detail. The advantages and properties of the proposed method are illustrated via synthetic and real case studies.

Original languageEnglish (US)
Pages (from-to)2071-2086
Number of pages16
JournalInternational Journal of Geographical Information Science
Volume25
Issue number12
DOIs
StatePublished - Dec 1 2011
Externally publishedYes

Fingerprint

spatial data
Logistics
logistics
kriging
prediction
Labels
response behavior
entropy
Random variables
Entropy
indicator
family
method

Keywords

  • categorical data
  • geostatistics
  • GLMM
  • indicator kriging
  • logistic regression

ASJC Scopus subject areas

  • Information Systems
  • Geography, Planning and Development
  • Library and Information Sciences

Cite this

A multinomial logistic mixed model for the prediction of categorical spatial data. / Cao, Guofeng; Kyriakidis, Phaedon C.; Goodchild, Michael.

In: International Journal of Geographical Information Science, Vol. 25, No. 12, 01.12.2011, p. 2071-2086.

Research output: Contribution to journalArticle

@article{ef4d07c0269e407f918e02537967680b,
title = "A multinomial logistic mixed model for the prediction of categorical spatial data",
abstract = "In this article, the prediction problem of categorical spatial data, that is, the estimation of class occurrence probability for (target) locations with unknown class labels given observed class labels at sample (source) locations, is analyzed in the framework of generalized linear mixed models, where intermediate, latent (unobservable) spatially correlated Gaussian variables (random effects) are assumed for the observable non-Gaussian responses to account for spatial dependence information. Within such a framework, a spatial multinomial logistic mixed model is proposed specifically to model categorical spatial data. Analogous to the dual form of kriging family, the proposed model is represented as a multinomial logistic function of spatial covariances between target and source locations. The associated inference problems, such as estimation of parameters and choice of the spatial covariance function for latent variables, and the connection of the proposed model with other methods, such as the indicator variants of the kriging family (indicator kriging and indicator cokriging) and Bayesian maximum entropy, are discussed in detail. The advantages and properties of the proposed method are illustrated via synthetic and real case studies.",
keywords = "categorical data, geostatistics, GLMM, indicator kriging, logistic regression",
author = "Guofeng Cao and Kyriakidis, {Phaedon C.} and Michael Goodchild",
year = "2011",
month = "12",
day = "1",
doi = "10.1080/13658816.2011.600253",
language = "English (US)",
volume = "25",
pages = "2071--2086",
journal = "International Journal of Geographical Information Science",
issn = "1365-8816",
publisher = "Taylor and Francis Ltd.",
number = "12",

}

TY - JOUR

T1 - A multinomial logistic mixed model for the prediction of categorical spatial data

AU - Cao, Guofeng

AU - Kyriakidis, Phaedon C.

AU - Goodchild, Michael

PY - 2011/12/1

Y1 - 2011/12/1

N2 - In this article, the prediction problem of categorical spatial data, that is, the estimation of class occurrence probability for (target) locations with unknown class labels given observed class labels at sample (source) locations, is analyzed in the framework of generalized linear mixed models, where intermediate, latent (unobservable) spatially correlated Gaussian variables (random effects) are assumed for the observable non-Gaussian responses to account for spatial dependence information. Within such a framework, a spatial multinomial logistic mixed model is proposed specifically to model categorical spatial data. Analogous to the dual form of kriging family, the proposed model is represented as a multinomial logistic function of spatial covariances between target and source locations. The associated inference problems, such as estimation of parameters and choice of the spatial covariance function for latent variables, and the connection of the proposed model with other methods, such as the indicator variants of the kriging family (indicator kriging and indicator cokriging) and Bayesian maximum entropy, are discussed in detail. The advantages and properties of the proposed method are illustrated via synthetic and real case studies.

AB - In this article, the prediction problem of categorical spatial data, that is, the estimation of class occurrence probability for (target) locations with unknown class labels given observed class labels at sample (source) locations, is analyzed in the framework of generalized linear mixed models, where intermediate, latent (unobservable) spatially correlated Gaussian variables (random effects) are assumed for the observable non-Gaussian responses to account for spatial dependence information. Within such a framework, a spatial multinomial logistic mixed model is proposed specifically to model categorical spatial data. Analogous to the dual form of kriging family, the proposed model is represented as a multinomial logistic function of spatial covariances between target and source locations. The associated inference problems, such as estimation of parameters and choice of the spatial covariance function for latent variables, and the connection of the proposed model with other methods, such as the indicator variants of the kriging family (indicator kriging and indicator cokriging) and Bayesian maximum entropy, are discussed in detail. The advantages and properties of the proposed method are illustrated via synthetic and real case studies.

KW - categorical data

KW - geostatistics

KW - GLMM

KW - indicator kriging

KW - logistic regression

UR - http://www.scopus.com/inward/record.url?scp=84859145627&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859145627&partnerID=8YFLogxK

U2 - 10.1080/13658816.2011.600253

DO - 10.1080/13658816.2011.600253

M3 - Article

VL - 25

SP - 2071

EP - 2086

JO - International Journal of Geographical Information Science

JF - International Journal of Geographical Information Science

SN - 1365-8816

IS - 12

ER -