异常波动条件下期货市场流动性的信息内涵研究——基于高阶矩的视角

The abnormal volatilities which often occur in futures markets have been paid much attention. The liquidity under abnormal volatilities, which are different from normal volatilities, also often represents abnormal returns and risk characteristics in futures markets. In order to discover the economic rationale and its internal rule behind the abnormal volatilities, this project investigates the information contents of liquidity under abnormal volatilities condition in futures markets from high moments viewpoints. The framwork of this study are as follows: The project firstly analyze the abnormal volatilities, high moments and their relationships. Then, the mechanism of information transmission between liquidity and high moments in futures markets under abnormal volatilities condition is investigated systematically in theory. And then, the time series models between liquidities, which are set up by market width and market depth aspects, and high moments variables are given. Based on the analysis of basic characteristics in variables of liquidity, high moments, and other variables by using high- and low-frequency data, we investigate empirically the internel relations between liquidity and high moments under abnormal volatilities in domestic and overseas futures markets respectively. Lastly, the liquidity management strategies in China under abnormal volatilities conditon are given pertinently according to the theoretical and empirical results. The research above can not only make up the shortage of microstructure but also make us know, hold and apply the essence and internal rule and more external and exacter in futures markets.

期货市场常出现异常波动，并备受关注。不同于正常波动，异常波动条件下期货市场的流动性往往表现出异常的收益和风险特征。为揭示异常波动背后的经济原理及其内在规律，将从高阶矩出发来探寻异常波动条件下期货市场流动性的信息内涵。为此，本课题首先对异常波动和高阶矩及其关系进行解析；接着从理论角度分析异常波动条件下期货市场流动性与高阶矩之间的信息传导机制；进而构建异常波动条件下基于宽度和深度角度的流动性与高阶矩变量之间的时间序列模型；然后，在对基于高频和低频数据的流动性和高阶矩等基本统计特征进行分析的基础上，系统对在异常波动条件下国内外期货市场的流动性与高阶矩变量之间的内在关系进行实证研究；最后，根据理论与实证分析结果，有针对性地给出中国期货市场异常波动条件下的流动性管理策略。对以上问题的研究不仅能够弥补期货市场微观结构研究的不足，也使得我们对期货市场本质及其内在规律的认识、把握和应用更为准确和客观。

对于中国的商品期货市场而言，异常波动现象仍时有发生，流动性作为市场微观结构理论中最重要的综合性指标，其内在的信息含量是巨大的。因此对异常波动下中国商品期货市场流动性的研究是十分必要和有意义的。. 本文首先分别通过SVCJ和MRS-SGED模型，识别出中国期货市场的异常波动，并比较两种模型识别的效果；然后，在此基础上研究流动性指标及其决定变量的统计特征；再者，用MIS模型选择出最合适的流动性指标——买卖价差；其次，通过回归模型研究了流动性指标与其决定变量的关系；最后，根据实证研究的结果为投资者和监管者分别提出建议。. 通过分析，本文发现以下重要结论：在正常和异常波动条件下，无论是流动性还是其决定要素，都表现出了不同程度的非对称性。在异常波动条件下，换手率、偏度和峰度对流动性的解释能力上升。在区分好坏消息的情况下，偏度的差别不是很明显，但峰度在坏消息更容易呈现“尖峰厚尾”的特点。. 最后，我们对异常波动下流动性管理的策略进行了研究，分别对投资者和监管者提出建议。对于投资者（主要是职业基金经理人）来说，在异常波动时期，应及早地调整投资组合配置较高的流动性好的资产，并且更加注重高阶矩风险，将对于金融资产高阶矩的计算纳入到投资组合的构建当中。对于监管者来说，应该考虑引入做市商制度，逐步放开涨跌停板限制来促进期货市场的流动性，同时注重资本账户开放过程中对于国内市场流动性的冲击。. 本文的创新点在于对异常波动下流动性和高阶矩的传导机制进行了研究并建立了相应的理论模型，此外还根据实证结果对期货市场的投资者和监管者分别提出合理化建议。