Geographically weighted regression in geospatial analysis
Loading...
Date
2012
Journal Title
Journal ISSN
Volume Title
Publisher
Springer Japan
Abstract
Geographically weighted regression (GWR) is a local spatial statistical technique for exploring spatial non-stationarity. The assumption in GWR is that observations nearby have a greater influence on parameter estimates than observations at a greater distance. This is very close to Tobler's first law of geography-everything is related to everything else, but near things are more related than distant things (Tobler 1970). GWR was developed on the basis of the traditional regression framework which incorporates local spatial relationships into the framework in an intuitive and explicit manner (Brunsdon et al. 1996; Fotheringham and Brunsdon 1999; Fotheringham et al. 2002). © 2012 Springer Japan. All rights reserved.
Description
Full text.
Keywords
Geospatial analysis, Geographically weighted regression, Spatial statistics, Spatial relationships
Citation
Thapa, R. B., & Estoque, R. C. (2012). Geographically weighted regression in geospatial analysis. Progress in Geospatial Analysis, 9784431540007, 85 - 96. https://www.researchgate.net/profile/Rajesh-Thapa-4/publication/261296287_Progress_in_Geospatial_Analysis/links/5cc930e492851c8d22106171/Progress-in-Geospatial-Analysis.pdf
Collections
Total visits
| views | |
|---|---|
| Geographically weighted regression in geospatial analysis | 8 |