In this paper, the authors construct country-specific chronologies of the house price bubbles for 12 OECD countries over the period 1969:Q1–2009:Q4. These chronologies are obtained using a combination of a fundamental approach and a filter approach. The resulting speculative bubble chronology is the one which provides the highest concordance between these two techniques. In addition, the authors suggest an early warning system based on three alternative approaches: A signalling approach, a logit model, and a probit model. It is shown that the latter two models allow much more accurate predictions of house price bubbles than the signalling approach. Furthermore, the predictive accuracy of the logit and probit models is high enough to make them useful in forecasting future speculative bubbles in housing market. Thus, this method can be used by the policymakers in their attempts to quickly detect house price bubbles and attenuate their devastating effects on the domestic and world economy.
The data set for this article can be found at: http://hdl.handle.net/1902.1/20452