Estimating the Marginal Implicit Price of the Spatially Diverse Environmental Amenities from Urban Open Space in Onondaga County, New York : A Geographically Weighted Regression Approach

This research applies Geographically Weighted Regression (GWR) to investigate how different types of urban open space and their spatial configurations are valued by residential property owners. To reflect residential property owners' perception of neighborhood and spatially heterogeneous configuration of urban open spaces, environmental amenity variables are measured at 100m and 1km radius buffer surrounding each residential property. GWR and Ordinary Least Squares (OLS) models are applied to analyze 4469 arms-length residential property transactions from Onondaga County, NY (USA) between Jan. 2000 and Dec. 2000 and are evaluated in terms of model fitting and spatial effects assessment. Results of this research show that GWR models outperform OLS models in model fitting and detecting spatial effects in model residuals, as well as is able to find spatial non-stationary relationships in larger number of environmental amenity variables. This research innovates in several ways. First, it innovates by considering spatial configuration of open spaces reflecting residential property owners' perspective. Second, this research illustrated spatially varying effect of urban open space variables using the marginal implicit price of each variable