Minimax Rate-optimal Estimation of High-dimensional Covariance Matrices with Incomplete Data.
Minimax Rate-optimal Estimation of High-dimensional Covariance Matrices with Incomplete Data.
复制标题
DOI:
10.1016/j.jmva.2016.05.002
复制
发表时间:
2016-09
影响因子:
1.6
通讯作者:
Zhang, Anru
中科院分区:
文献类型:
--
作者:
Cai, T. Tony;Zhang, Anru
关键词:
Missing data occur frequently in a wide range of applications. In this paper, we consider estimation of high-dimensional covariance matrices in the presence of missing observations under a general missing completely at random model in the sense that the missingness is not dependent on the values of the data. Based on incomplete data, estimators for bandable and sparse covariance matrices are proposed and their theoretical and numerical properties are investigated. Minimax rates of convergence are established under the spectral norm loss and the proposed estimators are shown to be rate-optimal under mild regularity conditions. Simulation studies demonstrate that the estimators perform well numerically. The methods are also illustrated through an application to data from four ovarian cancer studies. The key technical tools developed in this paper are of independent interest and potentially useful for a range of related problems in high-dimensional statistical inference with missing data.
登录
查看更多内容
影响因子:
4.5
作者:
Cai, Tony;Yuan, Ming
通讯作者:
Yuan, Ming
影响因子:
1.5
作者:
Lounici, Karim
通讯作者:
Lounici, Karim
影响因子:
2
作者:
Andreopoulos B;Anastassiou D
通讯作者:
Anastassiou D
影响因子:
2
作者:
Cai T;Ma Z;Wu Y
通讯作者:
Wu Y
影响因子:
4.5
作者:
Cai, T. Tony;Zhou, Harrison H.
通讯作者:
Zhou, Harrison H.