A new spatial-attribute weighting function for geographically weighted regression
نویسندگان
چکیده
In recent years, geographically weighted regression (GWR) has become popular for modeling spatial heterogeneity in a regression context. However, the current weighting function used in GWR only considers the geographical distances of trees in a stand, while the attributes (e.g., tree diameter) of the neighboring trees are totally ignored. In this study, we proposed a new weighting function that combines the “geographical space” and “attribute space” between the subject tree and its neighbors, such that (1) neighbors with greater geographical distances from the subject tree are assigned smaller weights, and (2) at a given geographical distance, neighboring trees with sizes that are similar to that of the subject tree are assigned larger weights. The results indicate that the GWR model with the new spatialattribute weighting function performs better than the one with the spatial weighting function in terms of model residuals and predictions for different spatial patterns of tree locations. Résumé : Dans les dernières années, la régression géographiquement pondérée (RGP) a souvent été utilisée pour modéliser l’hétérogénéité spatiale dans un contexte de régression. Toutefois, les fonctions de pondération actuellement disponibles pour la RGP considèrent seulement la distance entre les arbres dans un peuplement, alors que sont totalement ignorés les attributs des arbres voisins comme le diamètre. Dans la présente étude, nous proposons une nouvelle fonction de pondération qui combine l’information spatiale à celle d’attributs de l’arbre étudié et de ses voisins, de telle sorte (1) que les voisins les plus éloignés reçoivent une pondération plus faible et (2) qu’à une distance donnée, les voisins de taille similaire à l’arbre étudié reçoivent une pondération plus élevée. Les résultats montrent qu’une RGP utilisée avec la nouvelle fonction de pondération proposée a une meilleure performance que celle qui utilise un fonction de pondération uniquement spatiale, que ce soit en termes de résidus ou de valeurs prédites pour différents patrons de distribution spatiale des arbres. [Traduit par la Rédaction] Shi et al. 1005
منابع مشابه
Determining Effective Factors on Land Surface Temperature of Tehran Using LANDSAT Images And Integrating Geographically Weighted Regression With Genetic Algorithm
Due to urbanization and changes in the urban thermal environment and since the land surface temperature (LST) in urban areas are a few degrees higher than in surrounding non-urbanized areas, identifying spatial factors affecting on LST in urban areas is very important. Hence, by identifying these factors, preventing this phenomenon become possible using general education, inserting rules and al...
متن کاملModeling of the Relationships Between Spatio-Temporal Changes of Traffic Volume and Particulate Matter-2.5 Pollutant Concentration Based on Geographically Weighted Regression (GWR) and Inverse Distance Weighting (IDW) Model: A Case Study in Tehran M
Background and Aim: High concentrations of particulate matter-25 (PM2.5) have been the cause of the unhealthiest days in Tehran, Iran in recent years. This study was conducted with the aim of the spatio-temporal analysis of traffic volume and its relationship with PM2.5 pollutant concentrations in Tehran metropolis, Tehran during 2015-2018, using the Geographic Information System (GIS). Materi...
متن کاملComparison of Geographically Weighted Regression and Regression Kriging to Estimate the Spatial Distribution of Aboveground Biomass of Zagros Forests
Aboveground biomass (AGB) of forests is an essential component of the global carbon cycle. Mapping above-ground biomass is important for estimating CO2 emissions, and planning and monitoring of forests and ecosystem productivity. Remote sensing provides wide observations to monitor forest coverage, the Landsat 8 mission provides valuable opportunities for quantifying the distribution of above-g...
متن کاملComparison of the Performance of Geographically Weighted Regression and Ordinary Least Squares for modeling of Sea surface temperature in Oman Sea
In Marine discussions, the study of sea surface temperature (SST) and study of its spatial relationships with other ocean parameters are of particular importance, in such a way that the accurate recognition of the SST relationships with other parameters allows the study of many ocean and atmospheric processes. Therefore, in this study, spatial relations modeling of SST with Surface Wind Speed (...
متن کاملGeographically Weighted Regression: A Method for Exploring Spatial Nonstationarity
model which allows diferent relationships to exist at diferent points in space. This technique is loosely based on kernel regression. The method itself is introduced and related issues such as the choice of a spatial weighting function are discussed. Following this, a series of related statistical tests are considered which can be described generally as tests f o r spatial nonstationarity. Usin...
متن کامل