Signal Extraction with Kalman Filter: A Study of the Hong Kong Property Price Bubbles

Since Flood and Garber (1980), the debate surrounding speculative bubbles has never subsided. A key obstacle to resolve this issue is the identification problem. A bubble is usually inferred from some assumed fundamental determinants of a price. These assumptions could be over-simplified. Furthermore, there might be data measurement errors. In this paper, we attempt to capture such errors with a latent state variable. This variable is extracted with Kalman filter. Based on our empirical comparisons, we find that it is possible to attribute the observed large price swings in the property market of Hong Kong during the 1980s and 1990s to a periodically collapsing rational speculative bubble.