Jump robust daily covariance estimation by disentangling variance and correlation components

详细信息	查看全文 | 推荐本文 |

• 作者：Kris Boudt^a ; ^b ; ^{kris.boudt@econ.kuleuven.be} ; Jonathan Cornelissen^a ; ^{jonathan.cornelissen@econ.kuleuven.be} ; Christophe Croux^a ; ^{christophe.croux@econ.kuleuven.be}
• 关键词：Epps effect ; High-frequency data ; Integrated covariance ; Jumps ; Non-synchronous trading ; Realized covariance
• 刊名：Computational Statistics and Data Analysis
• 出版年：2012
• 期刊代码：19_01679473
• 类别：cp
• 出版时间：November, 2012
• 卷：56
• 期：11
• 页码：2993-3005
• 文件大小：266 K

摘要

A jump robust positive semidefinite rank-based estimator for the daily covariance matrix based on high-frequency intraday returns is proposed. It disentangles covariance estimation into variance and correlation components. This allows us to account for non-synchronous trading by estimating correlations over lower sampling frequencies. The efficiency gain of disentangling covariance estimation and the jump robustness of the estimator are illustrated in a simulation study. In an application to the Dow Jones Industrial Average constituents, it is shown that the proposed estimator leads to more stable portfolios.