Spatial analysis of housing price using geographically weighted regression (A case study in District 2 of Tehran Metropolitan City, Iran)

Housing is one of the most important bases for providing households' biological, economic, and social needs, and regarding the high volatility in supply and demand, a decent planning is required to pro-mote the quality of the residential environment. Housing prices are amongst those housing indices that could not be completely controlled by planners, but using spatial analysis of housing market performance and promoting the efficiency of plans and providing strategies and policies for housing plannings can improve the cotrolablity of the prices. The fluctuations in housing prices are one of the main urban management' challenges facing Tehran Metropolitan City. So, this paper is trying to investigate the spatial distribution of housing prices and to identify its determinants. One of the reasons for choosing district 2 of Tehran as the case in this research is the geographical spread of this district, which covers the center to the most northern urban areas of Tehran, and consequently includes a veraity of building types, residential patterns and housing prices. The recorded sales data in the Real Estate Market System, for apartment buildings in the district sold during Shahrivar and Mehr 1396, were used and applied in OLS and GWR regression techniques for modeling and analyzing the housing prices. In addition to identifying the variables affecting the housing price, the results indicate the utility of the GWR in comparison with the OLS technique in explaining the housing prices.