Forecast the Collapse of Speculative Bubbles for European Securitized Real Estate

Securitized real estate prices have historically shown tendencies to rise substantially and then fall quickly, which is the case with the prices of many other traded assets. These episodes of price rallies and crashes may constitute the presence of positive or negative speculative bubbles in the securitized real estate market. This has implications for real estate research: are these bubbles the result of irrational behaviour, are they predictable, how can investors formulate strategy? The present study focuses on bubbles in securitized real estate defined as the deviations of actual prices from fundamentals. It has two objectives: The first objective is to identify bubbles in the securitized market. The detection of stock bubbles has been the main focus of the relevant literature. Second the paper is concerned with timing the bubble. Of course timing the bubble cannot be accurate. However the investor would be able to use signals for buy or sell decisions. For this purpose we use the bubble detection models to forecast. We then develop trading rules based on inferences about the conditional probability of a crash and a rally and estimate risk adjusted returns from exercising these trading rules. We make use of a model specifically developed to detect periodically collapsing speculative bubbles by van Norden and Schaller (1999). This methodology is characterized by a time-varying probability of collapse of positive and negative speculative bubbles. The measures to calculate fundamental values in this study are (i) the dividend multiple approach and (ii) Campbell and Shillerís (1987) VaR model of dividend component of prices. The data employed in this study are selected REIT indices in Europe.