A0570
Title: Predicting the unpredictable: Ex-ante accuracy assessment of house price dispersion under unanticipated shock scenarios
Authors: Adam Chwila - University of Economics in Katowice (Poland)
Monika Hadas-Dyduch - University of Economics in Katowice (Poland)
Malgorzata Krzciuk - University of Economics in Katowice (Poland)
Tomasz Stachurski - University of Economics in Katowice (Poland) [presenting]
Alicja Wolny-Dominiak - University of Economics in Katowice (Poland)
Tomasz Zadlo - University of Economics in Katowice (Poland)
Abstract: "Prediction is very difficult, especially about the future," as Niels Bohr famously remarked. This challenge is particularly acute in the case of the prediction of housing market volatility. This market is extremely sensitive to fluctuations caused by political, environmental, technological, or social disturbances. The traditional ex-post approach does not take into consideration unforeseen events affecting a market. An improved framework is introduced for evaluating the ex-ante prediction accuracy of house price dispersion under unanticipated shocks. The proposed method uses a parametric bootstrap under a misspecified model, which enables the simulation of future price values and estimation of prediction errors across hypothetical shock scenarios. The analysis covers both classical (e.g., standard deviation) and quantile-based measures of dispersion applied to U.S. real estate transactions from 2015 to 2023. Future realizations of the variable are generated via linear mixed models that can capture spatial and temporal dependencies. The results may be of great importance for the practice of management in the real estate sector and decision-making because they provide an ex-ante prediction accuracy evaluation by simulating future realizations under specified shock scenarios.
