GWR

This study compares both non-spatial and spatial regression models to examine the factors associated with housing prices based on the municipal housing appraisal and real estate datasets for the city of Quito, Ecuador. The results suggest that compared to the traditional regression model, the GWR model is more effective at capturing housing market variations on a fine scale. Moreover, it reveals interesting findings on the spatial varying, sometimes opposite effects of some housing attributes on housing prices in different areas of the city, suggesting the potential impact from segregation.