Gradient Boosting–Based Machine Learning Methods in Real Estate Market Forecasting

DOI

10.2991/aisr.k.201029.039How to use a DOI?

Keywords

real estate market analysis, forecasting, housing market, housing, real estate, residential real estate market value

Abstract

Several approaches can be used to estimate the value of residential real estate. The sales comparison approach requires assessing a number of comparable residential properties and determining the degree of their compatibility. The cost approach requires using the information on all construction costs. The income approach requires a large amount of data on the market capacity, operation cost, expected operating expenses, and competitive opportunities. For the end customer or buyer and often for appraisers and realtors as well, these methods would involve processing a considerable amount of information. The sales comparison approach is used more frequently, since sufficient data for other approaches might not be publicly available. Nevertheless, all these methods are quite complex, and using them to estimate the value of a residential property can be time-consuming if performed without automated valuation models (AVMs). In the paper, the method of data collection is described, and the analysis based on these data is carried out. Moreover, the housing affordability index is determined (shows the number of years required to purchase a residential property). Finally, the most appropriate forecasting method with the least error is chosen, and the parameters of residential properties are determined and ranked according to the degree of their impact on the price.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Nikita Fedorov
AU  - Yulia Petrichenko
PY  - 2020
DA  - 2020/11/10
TI  - Gradient Boosting–Based Machine Learning Methods in Real Estate Market Forecasting
BT  - Proceedings of the 8th Scientific Conference on Information Technologies for Intelligent Decision Making Support (ITIDS 2020)
PB  - Atlantis Press
SP  - 203
EP  - 208
SN  - 1951-6851
UR  - https://doi.org/10.2991/aisr.k.201029.039
DO  - 10.2991/aisr.k.201029.039
ID  - Fedorov2020
ER  -