Forecasting House Prices in Germany
In the academic debate there is a broad consensus that house price fluctuations have a substantial impact on financial stability and real economic activity. Therefore, it is important to have timely information on actual and expected house price developments. The aim of this paper is to measure the latest price movements in different real estate markets in Germany and forecast near-term price developments. Therefore we construct hedonic house price indices based on real estate advertisements on the internet platform ImmobilienScout24. Then, starting with a naive AR(p) model as a benchmark, we investigate whether VAR and ARDL models using additional macroeconomic information can improve the forecasting performance as measured by the mean squared forecast error (MSFE). While these models reduce the forecast error only slightly, forecast combination approaches enhance the predictive power considerably.