【主题】Detecting Common Bubbles in a Large-Dimensional Financial System
【报告人】陈烨 (副教授, 首都经贸大学)
【摘要/Abstract】Asset price bubbles tend to occur simultaneously in multiple assets. Such co movement is likely to be driven by certain common latent factors in the market. Can we detect the presence of such common factors at an early stage of market exuberance? To answer this question, we build a factor model that includes both I(1) and explosive factors. The I(1) factor captures the driving force of market fundamentals. The explosive factor captures the driven forces of asset price bubbles, and is allowed to exist for parts of the sample period. We provide an algorithm for testing the presence and date stamping the origination of common bubbles in a large dimensional system. Asymptotics of the bubble test statistic are provided under both the null of no common bubbles and the alternative when there is a common bubble in the market. We show the consistency of the estimated bubble origination date. Simulation studies show satisfactory performance of the testing procedure in finite samples. Our methods are applied to the real estate markets of 89 major cities in China over the period of January 2003 to March 2013. Results suggest the presence of three common bubble episodes in Tier 1 and Tier 2 cities over the sample period, whereas little evidence of common bubbles is found in Tier 3 cities.