Using spatial randomisations to improve the utility of Geographically Weighted Regression model results

Access & Terms of Use
metadata only access
Abstract
Geographically Weighted Regression (GWR) and its variants are analysis methods that can cope with the multi-scale, spatially non-stationary relationships common to spatial data. They achieve this by using geographical sub-samples of the data for which one expects the complexity of any relationships to be simpler than over the whole study area.
Persistent link to this record
DOI
Link to Publisher Version
Author(s)
Laffan, Shawn
Bickford, S
Supervisor(s)
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2005
Resource Type
Conference Paper
Degree Type
UNSW Faculty