Capturing Housing Market Segmentation: An Alternative Approach based on Neural Network Modelling

摘要: Various location specific attributes cause segmentation of the housing market into submarkets. The question is, whether most relevant partitioning criteria are directly related to transaction price or other, socio-economic and physical, features location. On empirical side, several methods have been proposed that might be able capture this influence. This paper examines one these methods: neural network modelling with an application Helsinki, Finland. exercise shows how it is possible identify various dimensions submarket formation by uncovering patterns in dataset, also classification abilities two techniques: self-organising map (SOM) learning vector quantisation (LVQ). In clearly depends on factors: relative house type. Price-level has a smaller role respect.