郝红霞, 林金官, 汪红霞. 杠杆效应检验的一种新方法[J]. 应用概率统计, 2019, 35(5): 453-468. DOI: 10.3969/j.issn.1001-4268.2019.05.002
引用本文: 郝红霞, 林金官, 汪红霞. 杠杆效应检验的一种新方法[J]. 应用概率统计, 2019, 35(5): 453-468. DOI: 10.3969/j.issn.1001-4268.2019.05.002
HAO Hongxia, LIN Jinguan, WANG Hongxia. A New Method for Testing Leverage Effect[J]. Chinese Journal of Applied Probability and Statistics, 2019, 35(5): 453-468. DOI: 10.3969/j.issn.1001-4268.2019.05.002
Citation: HAO Hongxia, LIN Jinguan, WANG Hongxia. A New Method for Testing Leverage Effect[J]. Chinese Journal of Applied Probability and Statistics, 2019, 35(5): 453-468. DOI: 10.3969/j.issn.1001-4268.2019.05.002

杠杆效应检验的一种新方法

A New Method for Testing Leverage Effect

  • 摘要: 杠杆效应经常出现在金融风险管理、投资组合及期权定价等众多研究领域中. 然而, 实际数据中是否确实存在杠杆效应还有待考察.基于局部多项式估计和Kolmogorov-Smirnov非参检验方法,本文提出了一种新的杠杆效应的非参数检验方法, 构建了检验统计量,并得到了其渐近性质. 模拟研究表明, 本文提出的检验方法是有效的.采用P500数和MSFT数据进行了实证分析. 结果表明,P500指数和MSFT数据中确实存在杠杆效应, 这与金融领域中的观点相吻合.

     

    Abstract: Leverage effect often arises in many fields,such as financial risk management, portfolio and option pricing. However,it still remains to be studied that whether there is leverage effect or not in real data. Based on local polynomial regression estimation and Kolmogorov-Smirnov nonparametric test, this paper introduces a new nonparametric test statistic for the leverage effect, and some asymptotic properties are also presented. Simulation studies show that the proposed method performs well. Finally, empirical studies on SP500 index and Microsoft data imply that leverage effect exists in the real data, which is consistent with the idea in finance.

     

/

返回文章
返回