实时检验泡沫：基于严格局部鞅判别的泡沫检验方法

实时检验泡沫并对市场参与者和监管者进行预警，有利于防范和化解金融风险。基于检验泡沫的严格局部鞅判别原理，在RKSH方法的基础上拓展了一种倒向滚动检验方法（BSLM），并通过蒙特卡洛仿真和实证分析验证了新方法的有效性。研究发现：相对于传统严格局部鞅判别方法，BSLM方法对距离当前时刻较近的泡沫具有更高的检验势，并且能更早地事前预警泡沫；在对中国股票市场主要指数2000~2015年的泡沫检验中，新方法能判断泡沫的生成时间和破灭时间，发现上证指数和深圳成指的泡沫具有较强的联动性。

Real Time Detection of Financial Bubbles:A Method Based on Identification of Strict Local Martingale

Real time detection and early warning of bubble can help regulators and investors prevent and resolve financial risks. This paper develops a backward rolling window technique to detect exuberance in asset price series based on identification of strict local martingale (BSLM). Then, it verifies the effectiveness of the BSLM method by using Monte Carlo simulation and empirical analysis. The results show that compared with the traditional detecting method, BSLM can obtain more robust testing powers for detecting later bubble from current moment and give an earlier warning. Besides, when detecting the bubble of the main index of China’s stock market from 2000 to 2015, the BSLM method can effectively identify the origination and termination dates of bubbles and find that there exists a strong relation between the bubbles of Shanghai Composite Index and Shenzhen Component Index.