为了量化资产之间相依结构的局部特征,本文将小波阈值规则引入Copula参数估计,提出多元Copula密度的小波局部阈值估计量,发现Copula密度的光滑度指数、维数和采样容量是影响估值精度的重要因素,这一点也得到了以正态Copula为仿真算例的支持。本方法增强了参数Copula建模的局部自适应能力,进而有助于改进资产的市场风险估值与最优化配置。 In order to quantify the local characteristics of the dependency between assets, this paper introduces the wavelet threshold rule into the Copula parameter estimation, and proposes the local wavelet threshold estimate of the multivariate Copula density. It is found that the smoothness index, dimension and sampling capacity This is also supported by the normal Copula simulation example. This method enhances the local adaptive ability of the parametric Copula modeling, which helps to improve the market risk assessment and optimal allocation of assets.