Kentsel mekansal özelliklerin taşınmaz değerlemede kullanımı: CBS temelli bir model geliştirme önerisi

Thanks to technological developments in the field of GIS, it is possible to measure the quality of street networks. Given that, this study aims to create a model which examines the influence of urban qualities within the neighbourhood unit on housing prices. Since this study is limited to housing values, potential factors that affect housing value is discussed under two main headings: neighbourhood and building characteristics. The dataset was obtained by filtering the data from 151 valuation reports prepared by licensed appraisers. These are the reports of the residences for which an appraisal report has been prepared in the city of Izmir, Buca, in 2020. With the data obtained, a series of regression models were run via SPSS software. First, the influence of the building characteristics were tested, and the results showed that type (residence or not), size, presence of elevator, building quality, presence of parking lot, number of rooms were the significant factors on house values. Next, a reduced model is formed by eliminating the insignificant building characteristics and adding the neighbourhood characteristics to this reduced model to test the influence of neighbourhood characteristics on house values; after accounting for the important building characteristics. Results revealed that distances to the hospitals, green areas, global restaurants, shopping centers, university, and density of the groceries within the 400-meter buffer zone of housing were significant on estimating the value of housing. This study provides empirical evidence that data on neighbourhood characteristics leads better predictions of house values when it is not possible to collect a detailed data on building characteristics. The most important result of the study is that the housing value will be predictable with less detailed data.