Huff Inspired Gravity Model in Valuation of homes near Scenic lands -- A geographically weighted regression based hedonic model

• Published in: IDEAS Working Paper Series from RePEc
• Main Authors: Mittal, Jay, Byahut, Sweta
• Format: Paper
• Language: English
• Published: St. Louis Federal Reserve Bank of St. Louis 01.01.2019
• Subjects: Conservation easements

This research uses a hedonic Price modelling framework to assess the marginal implicit price effect of conservation easements (CE) lands on single family houses in Worcester, MA. The house price premium is anticipated with the growing visual accessibility from home to conservation easements lands. The CE lands of interest here are voluntarily protected, privately owned, scenic lands and are based in the urbanized area of City of Worcester, MA. The premium, and the visual accessibility was measured using the transaction of the surrounding homes, and homes spatial relationship with the CE lands. These protected CE lands are perpetually protected with natural, historic, and scenic characteristics that are attractive to the environmental amenity seekers. The home premiums as capitalized due to the visual accessibility of protected lands was measured using a combined weighted measure of ‘view’ and ‘proximity.' This was developed using the Huff's gravity model inspired index -- Gravity Inspired Visibility Index (GIVI). First, a detailed digital elevation model (DEM) raster with all view obstructing buildings and physicals structures stitched an the topography surface was generated and then the views and distances from homes to scenic lands were used to generate the GIVI, using the Viewshed analysis in ArcGIS. The geographically weighted regression (GWR) based hedonic model was then employed to measure the combined effect of both -- distance and view of scenic lands from each homes. Both the global (adjusted R sq =0.52, AICc =29,828) and the geographically weighted regression (GWR) models (adjusted R sq = 0.59, AICc =29,729) estimated the price effect, and the GWR model outperformed the global model. The results from the GWR model indicated an average 3.4% price premium on the mean value of homes in the study area. The spatial variation in home premiums (as percentage values) was also found clearer and more spatially clustered in the GWR model. The highest premium value fo