Bubble Analyses and Risk Management Implications for Real Estate Markets in US Cities.
| Autor: | LO, TING-YU, 羅廷玉 |
|---|---|
| Rok vydání: | 2019 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 107 This research differentiates the vast housing market literature, which typically used low-frequency monthly data, by adopting high-frequency daily house price data for 10 major US metropolitan areas to observe real estate bubble between 2001 and 2012. We decompose housing price returns into permanent fundamental and transitory fad components, and incorporate Markov-switching(MS) components into their volatility. Their dynamics are compared with those of naive (non-decomposed) housing price returns and the conventional Hodrick-Prescott (HP) method. Some interesting results yield informative implications for risk management. First, in terms of high and low volatility, high and low volatility phases are both evident for the 10 MSAs for the three models, particularly in the MS model. Differing from other viewpoints, the fundamental high volatility value is higher than the transitory fads. In the MS model, both high and low regimes of transitory fads are significant, suggesting regime-switching patterns characterize the fads of metropolitan housing returns, but the fundamental components do not have this feature. Second, the durative of high and low volatility phases in naive and HP models are qualitatively similar. In the MS model, the durative of transitory components are higher than the permanent fundamental components, suggesting that the formers are the main drivers of metropolitan housing price returns. Third, in general, the MS model suggests weaker bubble vulnerability than naive and HP models for the 10 MSAs based on their momentum reversal patterns. The 10 major US metropolitan areas’ housing price dynamics are primarily driven by transitory fads, which dominate fundamental components during the recent housing cycles. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|