Agent-based simulation of real-estate transactions

Abstract The prices of real-estate market influence the welfare of citizens and the business of real-estate investors. A well-known open challenge is to understand the repercussions of different combinations of individual buying/selling strategies on this market. The current approach is aimed at simulating these repercussions. For this purpose, a novel agent-based simulation tool includes common known strategies. This tool simulates the real-estate transactions from these strategies, showing the evolution of average prices and the results of each strategy (i.e. their success ratio, average price of their transactions and average waiting time). The underlying framework is extensible so that users can easily define and simulate new strategies. The experimentation of this work includes micro-validation of each kind of strategy. It also assesses this tool using a Spanish real-estate website (Idealista.com). The calibration of the tool was performed with the data of a small-medium city (Huesca), and the validation was performed for a different city (Teruel) to avoid overfitting. The results advocate that the current simulator might be an appropriate first step towards the simulation and analysis of the combination of certain buying and selling behaviors in the real-estate market.