Geographically weighted regression - modelling spatial non-stationarity

被引：938

作者：

Brunsdon, C ^{[1
]}

Fotheringham, S ^{[1
]}

Charlton, M ^{[1
]}

机构：

[1] Newcastle Univ, Dept Town & Country Planning, Newcastle Upon Tyne NE1 7RU, Tyne & Wear, England

来源：

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES D-THE STATISTICIAN | 1998年 / 47卷

关键词：

exploratory data analysis; geographically weighted regression; Monte Carlo testing; regression; spatial non-stationarity;

D O I：

10.1111/1467-9884.00145

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

In regression models where the cases are geographical locations, sometimes regression coefficients do not remain fixed over space. A technique for exploring this phenomenon, geographically weighted regression is introduced. A related Monte Carte significance test for spatial non-stationarity is also considered. Finally, an example of the method is given, using limiting long term illness data from the 1991 UK census.

引用

页码：431 / 443

页数：13

共 17 条

[1]

Cassetti E., 1972, GEOGR ANAL, V4, P432

[2]

CHATFIELD C, 1994, GUIDE STAT PROBLEM S

[3] LOCALLY WEIGHTED REGRESSION - AN APPROACH TO REGRESSION-ANALYSIS BY LOCAL FITTING [J].

CLEVELAND, WS ;

DEVLIN, SJ .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1988, 83 (403) :596-610

[4]

Dale A., 1993, The 1991 Census Users Guide

[5]

FOTHERINGHAM AS, 1992, ENV PLANNG A, V25, P156

[6]

FOTHERINGHAM AS, 1994, ENV PLANNG A, V23, P34

[7]

Goldstein H., 1987, MULTILEVEL MODELS ED

[8]

GORR WL, 1994, GEOGR ANAL, V26, P67

[9]

Holt D, 1996, GEOGR ANAL, V28, P244

[10]

HOPE ACA, 1968, J ROY STAT SOC B, V30, P582

← 1 2 →