Simulation of Residential Real Estate Markets in the Largest Russian Cities

Economy of Regions. 2022. Vol. 18. No. 2. P. 609–622.

Alekseev A.

The existing mass appraisal models and mathematical tools for predicting the market value of residential property have a number of disadvantages, as they are developed for individual regions. Without considering the constantly changing economic environment, these models quickly become outdated and require constant updating. Thus, they are not suitable for construction business optimisation. The study aims to create a universally applicable real estate appraisal system for Russian cities, regardless of the constantly changing economic situation. This goal was achieved through the creation of a neural network, whose input parameters include construction and operational data, geographical factors, time effect, as well as a number of indicators characterising the economic situation in specific regions, Russia and the world. In order to examine the dynamics of real estate markets in the Russian Federation, statistical data for neural network training were collected over a long period from 2006 to 2020. Virtual computer experiments were performed for testing the developed system. They showed that minimum size one-room apartments of 16 square meters have the highest unit cost per square meter in Moscow. Two-room apartments with an area of 90 square meters have the maximum price, as well as 100 sq. m. three-room, 110 sq. m. four-room and 120 sq. m. five-room apartments. In Ekaterinburg, two-room apartments with a total area of 30 square meters have the highest cost per square meter; the same applies for 110 sq. m. three-room, 130 sq. m. four-room and 150 sq. m. five-room apartment. Thus, the proposed system can be used to optimise the construction business. It can be also be useful for government institutions concerned with urban real estate market management, property taxation, and housing market improvement.

Keywords: нейронная сеть, real estate, прогнозирование цен, summary cost estimate, artifitial neural networks, городская недвижимость, прогнозирование, стоимость, рынок, мак-роэкономические факторы, нейронная сеть, Forecasting