Outlier Detection via a Block Diagonal Product Estimator

作     者：LI Chikun JIN Baisuo LI Chikun;JIN Baisuo

作者机构：School of ManagementUniversity of Science and Technology of ChinaHefei 230026China 

出 版 物：《Journal of Systems Science & Complexity》 (系统科学与复杂性学报（英文版）)

年 卷 期：2022年第35卷第5期

页      面：1929-1943页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 

基　　金：supported by the National Natural Science Foundation of China under Grant Nos.71873128 and 72111530199 

主　　题：Block diagonal high dimension minimum covariance determinant estimator 

摘      要：Outlier detection is a fundamental topic in robust *** outlier detection methods try to find a clean subset of given size,which is used to estimate the location vector and scatter matrix,and the outliers can be flagged by the Mahalanobis ***,methods such as the minimum covariance determinant approach cannot be applied directly to high-dimensional data,especially when the dimension of the sample is greater than the sample size.A novel fast detection procedure based on a block diagonal partition is proposed,and the asymptotic distribution of the modified Mahalanobis distance is *** authors verify the specificity and sensitivity of this procedure by simulation and real data analysis in high-dimensional settings.