GeoWeightedModel : An R-Shiny package for Geographically Weighted Models

Abstract

This paper describes GeoWeightedModel, a R package, which provides a graphical user friendly web application to perform techniques from a subarea of spatial Statistics known as Geographically Weighted (GW) models, such as Geographically Weighted Regression (GWR) and its extensions: Robust GWR, Generalized GWR, Heteroskedastic GWR, Mixed GWR, and “Scalable​ GWR), Geographically Weighted Principal Component Analysis, and Geographically Weighted Discriminant analysis. It also allows calculating a basic and robust Geographically weighted summary. The main goal of GeoWeightedModel package was to make the workflow easier to use, especially for those who are not familiar with the R environment. With GeoWeightedModel, analyses can be performed interactively (point-and-click way) in a web browser, making the applications easier for many more researchers. In addition with this tool, the results of the analyses can be mapped providing a valuable tool for exploring the spatial heterogeneity of the data.

Keywords

• Geographically weighted
• Geographically weighted analysis
• Spatial heterogeneity
• Geographically weighted regression
• Geographically weighted principal component analysis
• Discriminant geographically weighted analysis