• there is currently no "holy grail" for housing market segmentation
  • data-driven segmentation is more objective than a priori segmentation by humans
  • data-driven segmentation is mostly powered by clustering algorithms
  • these are also extended with additional approaches (geostatistics, different preprocessing, etc.)
There was already a huge effort spent to prove the existence of housing market segments, how to utilize them to improve valuation accuracy, and gain knowledge about the inner structure of the whole superior housing market. Accordingly, many different methods on the topic were explored, but there is still no universal framework known. The aim of this article is to review some previous studies on data-driven housing market segmentation methods with a focus on clustering methods and their ability to capture market segments with respect to the shape of clusters, fuzziness, and hierarchical structure.
This work is supported by the grant Segmentation of residential real estate market for property valuation purposes - empirical evidence and statistical modeling (ÚSI-K-21-6920), which is realized within the project Quality Internal Grants of BUT (KInG BUT), Reg. No. CZ.02.2.69 / 0.0 / 0.0 / 19_073 / 0016948, and ¬¬financed from the OP RDE.
Journals System - logo
Scroll to top