Volatility model applications in China's SSE50 options market
研究多种波动率模型在中国上证50指数期权中的表现，发现GARCH模型在样本内拟合和样本外预测中优于其他模型，并基于GARCH预测与隐含波动率的价差构建了盈利的交易策略。
Abstract We investigate the effectiveness of various volatility models using China's Shanghai Stock Exchange‐50 (SSE50) Index. Regarding in‐sample fit, the generalized autoregressive conditional heteroscedasticity (GARCH) and the variants of the GARCH model perform much better than the autoregressive conditional heteroscedasticity model. However, we do not observe any significant asymmetric volatility response to past returns in the GJR–GARCH model. For out‐of‐sample forecasting of the future realized volatility, in five of the seven options we investigate, the GARCH volatility forecast outperforms the option implied volatility. We formulate a trading strategy by exploiting the spread between the GARCH volatility forecast and the option implied volatility, and show robust profits when applied to the SSE50 options.