摘要
模拟蒙特卡洛分析发现,当两个泡沫间隔很短或第一个泡沫比第二个持续时间较长时,BSADF检验方法不能准确地估计出第二个泡沫,而本文修正的BSADF检验方法能有效地克服此问题。通过比较这两种方法对上证综指泡沫的产生和破灭时间的估计差异,发现BSADF检验只能检验出一个泡沫,泡沫持续期为2006年12月至2007年12月,而修正的BSADF发现了紧邻的第二个短暂性泡沫,发生在金融危机期间。
Monte Carlo simulation shows that, in the case of either short time interval between the two bubbles or longer duration of the first bubble than the second, BSADF based testing of stock price bubbles can not accurately estimate the second bubble, but the modified BSADF testing can efficiently tackle this problem. Furthermore, we compare the difference between BSADF and modified BSADF methods in identifying the starting and bursting of bubbles in Shanghai composite index. The results indicate that the BSADF can only tests out a bubble spanned from December 2006 to December 2007, while the modified BSADF can further find out the second neighbor bubble which occurred in the medium-term financial crisis between May and July in 2009.
出处
《数量经济技术经济研究》
CSSCI
北大核心
2012年第4期110-122,共13页
Journal of Quantitative & Technological Economics
关键词
修正BSADF
上证综指
泡沫
Modified BSADF
Shanghai Composite Index
Bubbles