2026-07-22T02:02:20Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:18935 2024-03-06T11:03:00Z ddc:004 open_access

Introducing a General Framework for Locally Weighted Spatial Modelling Based on Density Regression (Short Paper) Hu, Yigong Lu, Binbin Harris, Richard Timmerman, Richard Spatial heterogeneity Multidimensional space Density regression Spatial statistics Traditional geographically weighted regression and its extensions are important methods in the analysis of spatial heterogeneity. However, they are based on distance metrics and kernel functions compressing differences in multidimensional coordinates into one-dimensional values, which rarely consider anisotropy and employ inconsistent definitions of distance in spatio-temporal data or spatial line data (for example). This article proposes a general framework for locally weighted spatial modelling to overcome the drawbacks of existing models using geographically weighted schemes. Underpinning it is a multi-dimensional weighting scheme based on density regression that can be applied to data in any space and is not limited to geographic distance. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Yigong Hu and Binbin Lu and Richard Harris and Richard Timmerman 2023 Is Part Of LIPIcs, Volume 277, 12th International Conference on Geographic Information Science (GIScience 2023) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.GIScience.2023.40 urn:nbn:de:0030-drops-189354 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GIScience.2023.40 eng https://creativecommons.org/licenses/by/4.0/legalcode