Research project

The prime objective is to enlarge and enhance the current methodological framework of spatial hedonic house price models taking advantage of new technologies like Geographic Information Systems and airborne laser scanning in a consistent way and develop an integrative modeling framework. The project enhances the predictive accuracy of real estate pricing models by deriving local environmental neighborhood indicators based on highly accurate airborne laser scanning data. This is a crucial point because when conducting an analysis on a large-scale, some essential environmental determinants influencing the price are not available in traditional databases, actuated by traditional federal statistical offices. Hence, this data source is highly capable for large-scale modeling because it offers a highly detailed topographic representation of the environment and a highly accurate vertical representation of the structural component of the complex urban landscape. It is anticipated that variables like the average solar radiation hours per day dependent on aspect and occlusion of the flats, the density of (visible) vegetation in the immediate neighborhood, the percentage of horizon visible, among others, help to enhance traditional methodologies in applied econometrics and boosting their predictive power. For estimation purposes of hedonic price functions, this research uses two well-known state-of-the-art econometric modeling approaches: the spatial autoregressive model and the geographically weighted regression. The project addresses the local housing market of the third district of Vienna as a case study, estimating hedonic pricing models of flats.