Geographically weighted Poisson regression for disease association mapping

Tomoki Nakaya, Stewart Fotheringham, C. Brunsdon, M. Charlton

Research output: Contribution to journalArticle

203 Citations (Scopus)

Abstract

This paper describes geographically weighted Poisson regression (GWPR) and its semi-parametric variant as a new statistical tool for analysing disease maps arising from spatially non-stationary processes. The method is a type of conditional kernel regression which uses a spatial weighting function to estimate spatial variations in Poisson regression parameters. It enables us to draw surfaces of local parameter estimates which depict spatial variations in the relationships between disease rates and socio-economic characteristics. The method therefore can be used to test the general assumption made, often without question, in the global modelling of spatial data that the processes being modelled are stationary over space. Equally, it can be used to identify parts of the study region in which 'interesting' relationships might be occurring and where further investigation might be warranted. Such exceptions can easily be missed in traditional global modelling and therefore GWPR provides disease analysts with an important new set of statistical tools. We demonstrate the GWPR approach applied to a dataset of working-age deaths in the Tokyo metropolitan area, Japan. The results indicate that there are significant spatial variations (that is, variation beyond that expected from random sampling) in the relationships between working-age mortality and occupational segregation and between working-age mortality and unemployment throughout the Tokyo metropolitan area and that, consequently, the application of traditional 'global' models would yield misleading results.

Original languageEnglish (US)
Pages (from-to)2695-2717
Number of pages23
JournalStatistics in Medicine
Volume24
Issue number17
DOIs
StatePublished - Sep 15 2005
Externally publishedYes

Fingerprint

Poisson Regression
Tokyo
Mortality
Unemployment
Kernel Regression
Nonstationary Processes
Weighting Function
Random Sampling
Japan
Spatial Data
Segregation
Modeling
Economics
Estimate
Exception
Spatial Regression
Demonstrate
Relationships

Keywords

  • Geographically weighted regression
  • Kernel mapping
  • Poisson regression
  • Spatial analysis
  • Tokyo

ASJC Scopus subject areas

  • Epidemiology

Cite this

Geographically weighted Poisson regression for disease association mapping. / Nakaya, Tomoki; Fotheringham, Stewart; Brunsdon, C.; Charlton, M.

In: Statistics in Medicine, Vol. 24, No. 17, 15.09.2005, p. 2695-2717.

Research output: Contribution to journalArticle

Nakaya, Tomoki ; Fotheringham, Stewart ; Brunsdon, C. ; Charlton, M. / Geographically weighted Poisson regression for disease association mapping. In: Statistics in Medicine. 2005 ; Vol. 24, No. 17. pp. 2695-2717.
@article{c6bdf757e5674f0a898d656fe87cefe8,
title = "Geographically weighted Poisson regression for disease association mapping",
abstract = "This paper describes geographically weighted Poisson regression (GWPR) and its semi-parametric variant as a new statistical tool for analysing disease maps arising from spatially non-stationary processes. The method is a type of conditional kernel regression which uses a spatial weighting function to estimate spatial variations in Poisson regression parameters. It enables us to draw surfaces of local parameter estimates which depict spatial variations in the relationships between disease rates and socio-economic characteristics. The method therefore can be used to test the general assumption made, often without question, in the global modelling of spatial data that the processes being modelled are stationary over space. Equally, it can be used to identify parts of the study region in which 'interesting' relationships might be occurring and where further investigation might be warranted. Such exceptions can easily be missed in traditional global modelling and therefore GWPR provides disease analysts with an important new set of statistical tools. We demonstrate the GWPR approach applied to a dataset of working-age deaths in the Tokyo metropolitan area, Japan. The results indicate that there are significant spatial variations (that is, variation beyond that expected from random sampling) in the relationships between working-age mortality and occupational segregation and between working-age mortality and unemployment throughout the Tokyo metropolitan area and that, consequently, the application of traditional 'global' models would yield misleading results.",
keywords = "Geographically weighted regression, Kernel mapping, Poisson regression, Spatial analysis, Tokyo",
author = "Tomoki Nakaya and Stewart Fotheringham and C. Brunsdon and M. Charlton",
year = "2005",
month = "9",
day = "15",
doi = "10.1002/sim.2129",
language = "English (US)",
volume = "24",
pages = "2695--2717",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "17",

}

TY - JOUR

T1 - Geographically weighted Poisson regression for disease association mapping

AU - Nakaya, Tomoki

AU - Fotheringham, Stewart

AU - Brunsdon, C.

AU - Charlton, M.

PY - 2005/9/15

Y1 - 2005/9/15

N2 - This paper describes geographically weighted Poisson regression (GWPR) and its semi-parametric variant as a new statistical tool for analysing disease maps arising from spatially non-stationary processes. The method is a type of conditional kernel regression which uses a spatial weighting function to estimate spatial variations in Poisson regression parameters. It enables us to draw surfaces of local parameter estimates which depict spatial variations in the relationships between disease rates and socio-economic characteristics. The method therefore can be used to test the general assumption made, often without question, in the global modelling of spatial data that the processes being modelled are stationary over space. Equally, it can be used to identify parts of the study region in which 'interesting' relationships might be occurring and where further investigation might be warranted. Such exceptions can easily be missed in traditional global modelling and therefore GWPR provides disease analysts with an important new set of statistical tools. We demonstrate the GWPR approach applied to a dataset of working-age deaths in the Tokyo metropolitan area, Japan. The results indicate that there are significant spatial variations (that is, variation beyond that expected from random sampling) in the relationships between working-age mortality and occupational segregation and between working-age mortality and unemployment throughout the Tokyo metropolitan area and that, consequently, the application of traditional 'global' models would yield misleading results.

AB - This paper describes geographically weighted Poisson regression (GWPR) and its semi-parametric variant as a new statistical tool for analysing disease maps arising from spatially non-stationary processes. The method is a type of conditional kernel regression which uses a spatial weighting function to estimate spatial variations in Poisson regression parameters. It enables us to draw surfaces of local parameter estimates which depict spatial variations in the relationships between disease rates and socio-economic characteristics. The method therefore can be used to test the general assumption made, often without question, in the global modelling of spatial data that the processes being modelled are stationary over space. Equally, it can be used to identify parts of the study region in which 'interesting' relationships might be occurring and where further investigation might be warranted. Such exceptions can easily be missed in traditional global modelling and therefore GWPR provides disease analysts with an important new set of statistical tools. We demonstrate the GWPR approach applied to a dataset of working-age deaths in the Tokyo metropolitan area, Japan. The results indicate that there are significant spatial variations (that is, variation beyond that expected from random sampling) in the relationships between working-age mortality and occupational segregation and between working-age mortality and unemployment throughout the Tokyo metropolitan area and that, consequently, the application of traditional 'global' models would yield misleading results.

KW - Geographically weighted regression

KW - Kernel mapping

KW - Poisson regression

KW - Spatial analysis

KW - Tokyo

UR - http://www.scopus.com/inward/record.url?scp=24944468295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24944468295&partnerID=8YFLogxK

U2 - 10.1002/sim.2129

DO - 10.1002/sim.2129

M3 - Article

VL - 24

SP - 2695

EP - 2717

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 17

ER -