The aim of exploring and monitoring housing-market fundamentals (prices, dwelling features, area density, residents, and so on) on a macrolocational level relates to both public and private sector policymaking. Housing market segmentation (that is, the emergence of housing submarkets), a concept with increasing relevance, is defined as the differentiation of housing in terms of the income and preferences of the residents and in terms of administrative circumstances. In order to capture such segmentation empirically, the author applies a fairly new and emerging technique known as the ‘self-organising’ map (SOM), or ‘Kohonen map’. The SOM is a type of (artificial) neural network—a nonlinear and flexible (that is, nonparametric or semiparametric) regression and ‘machine learning’ technique. By utilising the ability of the SOM to visualise patterns, one can analyse various dimensions within the variation of the dataset. Segmentation may then be detected depending on the resulting patterns across the map layers, each of which represents the data variation for one input variable. Utilising an inductive modelling strategy, the author runs cross-sectional and nationwide data on the owner-occupied housing markets of Finland (documentation presented elsewhere), the Netherlands, and Hungary with the SOM technique. On the basis of the resulting configurations certain regularities (similarities and differences) across the three national contexts are identified. In all three cases the segments are determined by physical and institutional differences between the housing bundles and localities. The exercise demonstrates how the inductive SOM-based approach is well-suited for illustrating the contextual factors that determine housing market structure.
The potential of transformative video design for improving caregiver’s wellbeing