基于混合频率数据的大维协方差阵的估计及其应用  

Estimation and Application Study on the Large Dimensional Covariance Matrix of Mixed-Frequency Data

作　　者：刘丽萍[1] 

机构地区：[1]贵州财经大学数学与统计学院,贵阳550025

出　　处：《系统科学与数学》2017年第6期1532-1540,共9页Journal of Systems Science and Mathematical Sciences

基　　金：国家社会科学基金项目(16CTJ013);贵州省教育厅2015年度普通本科高校自然科学研究项目(黔教合KY字[2015]423);2015年全国统计科学研究项目(2015LY19)资助课题

摘　　要：大维数据给传统的协方差阵估计方法带来了巨大的挑战,数据维度和噪声的影响使得协方差阵的估计较为困难.在文章的研究中,将高频数据和低频数据相结合,提出了基于混合频率数据的协方差阵的估计和预测模型——MFD模型,MFD模型在解决了维数诅咒的同时还考虑了过去市场信息对协方差阵的影响,动态地估计和预测了未来的协方差阵.通过实证研究发现:较基于低频数据和高频数据的协方差阵估计和预测模型而言,MFD模型明显提高了高维协方差阵的估计和预测效率;并且将其应用在投资组合时,投资者获得了更高的投资收益和经济福利.High dimensional data poses great challenges to the traditional estimation of covariance, it is difficult to estimate the covariance matrix, because of the influence of data dimension and noise. In the study of this paper, we combine high frequency data and low frequency data, and put forward the estimation and prediction model of covariance matrix MFD model, which is based on the mixed frequency data. The MFD model not only solves the curse of dimensionality but also considers the impact of the past market information on the covariance matrix. The future covariance matrix is dynamic estimated and predicted. Through empirical studies, it is found that MFD model significantly improves the efficiency of estimation and prediction of large matrix and investors obtain higher returns and economical welfare when the MF model is applied in portfolio.

关 键 词：混合频率数据 MFD模型 大维协方差降 

分 类 号：F832.51[经济管理—金融学] O212.4[理学—概率论与数理统计]