Data-driven agent-based modeling of the Hungarian housing market - András Borsos
The Central Bank of Hungary developed a complex, modular, 1:1 scale, agent-based model of the Hungarian residential housing market, where all the 4 million households and their relevant characteristics are represented based on empirical micro‐level data. The model features transactions in the housing and rental markets, a construction sector, buy‐to‐let investors, housing loans, house price dynamics and a procyclical banking sector regulated by a macroprudential authority. After a brief elaboration of the features of this model, the talk will cover four applications:
- Optimal choice of scaling in economic agent-based models: trade-offs between runtime, accuracy, and precision;
- Comprehensive evaluation of borrower-based macroprudential policies;
- Evaluation of demand- and supply-side policy schemes supporting first-time home buyers;
- Interactions between the housing market and the macroeconomic environment using a SVAR extension of the ABM framework.