利用2015年2月9日至2021年6月23日1550个交易日的股票和期权市场数据，基于对广义无模型隐含波动率方法进行调整后的GVIX-EVT模型进行尾部风险测度，并将GVIX-EVT测度与传统尾部风险测度做对比，分析其对于极端风险事件造成的股票市场下行波动的预测效果。结果表明：期权市场价格相对历史价格对未来波动率变化所包含的信息更多，符合有效市场理论，且通过GVIX-EVT模型的尾部风险测度更有利于预测极端风险事件对我国股票市场的尾部损失程度；将隐含波动率与极值理论相结合得到的GVIX-EVT尾部风险测度，有助于风险管理者防范和应对极端风险事件对资本市场的冲击。

Adopting the stock and option market data of 1550 trading days from February 9, 2015 to June 23, 2021,the paper measures tail risk based on the GVIX-EVT model adjusted by the generalized model-free implied volatility method and compares the measurement of the GVIX-EVT with that of the traditional tail risk to analyze its prediction effect on the downward fluctuation of the stock market caused by extreme risk events. The results show that the option market price contains more information about the future volatility than the historical price, which is in line with the efficient market theory, and the tail risk measurement by the GVIX-EVT model is relatively more effective in predicting the tail loss of China’s stock market caused by extreme risk events. The tail risk measurement of GVIX-EVT obtained by combining implied volatility with extreme value theory is helpful for risk managers to prevent and tackle the impact of extreme risk events on the capital market.

F831.5