高维条件协方差矩阵的非线性压缩估计及其在构建最优投资组合中的应用

发布时间：2018-04-25 19:43

本文选题：非线性压缩 + 线性压缩　；参考：《中国管理科学》2017年08期

【摘要】：本文将非线性压缩方法运用到DCC和BEKK模型中,用非线性的压缩估计量代替MMLE估计中初始的样本协方差矩阵,大大提高了高维DCC和BEKK模型的估计效率,并突破性地使得横截面维度大于时间维度时,DCC和BEKK模型的有效估计成为可能。蒙特卡洛模拟发现:非线性压缩方法对于DCC和BEKK模型估计的优化作用显著,且优化程度随着横截面维度和时间维度的比值增大而增加。实证分析进一步说明了非线性压缩方法对于准确估计高维条件协方差矩阵、从而提高组合选择效率的重要作用。
[Abstract]:In this paper, the nonlinear compression method is applied to the DCC and BEKK models. The initial sample covariance matrix in the MMLE estimation is replaced by the nonlinear compression estimator, which greatly improves the estimation efficiency of the high-dimensional DCC and BEKK models. It also makes it possible to estimate the BEKK and DCC models when the cross section dimension is larger than the time dimension. Monte Carlo simulation shows that the nonlinear compression method plays an important role in the estimation of DCC and BEKK models, and the degree of optimization increases with the ratio of cross section dimension to time dimension. The empirical analysis further illustrates the importance of nonlinear compression method in estimating the high dimensional conditional covariance matrix accurately and improving the efficiency of combination selection.
【作者单位】：华中科技大学经济学院;
【基金】：国家自然科学基金面上资助项目(71671070)
【分类号】：F830.59;O212.1

【相似文献】