ABSTRACT

Recent advances in data disaggregation and the generation of synthetic spatial microdata are exploited for identifying local housing market attributes. Traditional public sector ‘small’ data is disaggregated to yield big (synthetic) spatial microdata using a three-stage approach. First, an allocation algorithm is used to attach synthetic socio-economic attributes to residential buildings. Second, inconsistencies between housing values and the ascribed socio-economic attributes of the resident population are identified. These indicate incipient clusters of housing market change. Third, clusters are typologized using the synthetic socio-economic microdata coupled with building attributes data. This yields information on housing market attributes such as segmentation and dynamics such as gentrification. The approach is operationalized for the entire stock of residential units and households in Israel and can be easily reproduced in other national contexts.