Option-implied Higher-order Co-moments: Extraction, Analysis and Trading Strategy
在多资产投资组合的分析框架中,除单资产的收益及波动率等高阶矩外,协偏度以及协峰度等高阶协矩亦是不可忽视的系统性风险度量。本文借鉴Bakshi等(2003)等文献的研究框架,利用台湾期权市场数据提取隐含高阶总矩、隐含协矩和隐含特质矩,探讨其各自对相关已实现矩的预测效果;并进一步构建协矩交易策略。结果表明:相较偏度及峰度,隐含波动率与实际波动率走势及统计特征均更为一致。协偏度、协峰度的波动相比协方差要剧烈得多。引入多市场信息的预测效果要优于单独采用某一市场信息的效果。协矩交易策略方面:历史矩与隐含矩信息在组合构建的差异主要体现在偏度与峰度等更高阶矩上。历史协方差与协偏度在市场趋稳时期表现相对较佳;隐含协矩的优势在于策略构建的稳健性更好。期权市场信息的有效反映取决于市场的成熟、演进及交易活跃度的提升。In multi-asset portfolio analysis framework, in addition to the single asset returns and volatility of such moments, the higher-order co-moments such as co-skewness and co-kurtosis are also systemic risk measure which cannot be ignored. In Bakshi etc. （2003） and other scholars＇ research framework, this paper extracts implied high-order total moments, implied co-moments and idiosyncratic moments with Taiwan options market data. This paper inspects their characteristic differences, discusses their respective prediction effect of related realized moments, and then builds co-moments trading strategies. The research results indicate that： In comparison of skewness and kurtosis, the tendency and statistical characteristics of the implied volatility and the actual volatility are more consistent. The fluctuation of co-skewness and co-kurtosis is much more severe than covariance. Co-moments trading strategies： The differences of historical moments and implied moments mainly reflecte in the higher-order co-moments such as co-skewness and co-kurtosis. The historical eovariance and coskewness strategies perform better in stabilization period. The advantage of implied co-moments is their robustness in trading strategies. The effective reflection of options market information depends on the market＇ s matures, evolution and ascension of trading activity.