L1 least squares for sparse high-dimensional LDA | |
Li, Yanfang ; Jia, Jinzhu | |
2017 | |
关键词 | High-dimensional LDA Lasso sparsity LINEAR DISCRIMINANT-ANALYSIS LASSO CLASSIFICATION REGRESSION |
英文摘要 | This paper studies high-dimensional linear discriminant analysis (LDA). First, we review the l(1) penalized least square LDA proposed in [10], which could circumvent estimation of the annoying high-dimensional covariance matrix. Then detailed theoretical analyses of this sparse LDA are established. To be specific, we prove that the penalized estimator is l(2) consistent in high-dimensional regime and the misclassification error rate of the penalized LDA is asymptotically optimal under a set of reasonably standard regularity conditions. The theoretical results are complementary to the results to [10], together with which we have more understanding of the l(1) penalized least square LDA (or called Lassoed LDA).; National Science Foundation of China [11101005, 11571021]; Key Lab of Mathematical Economics and Quantitative Finance (Ministry of Education); Key lab of Mathematics and Applied Mathematics (Ministry of Education); MOE-Microsoft Key Laboratory of Statistics and Information Technology of Peking University; SCI(E); ARTICLE; 1; 2499-2518; 11 |
语种 | 英语 |
出处 | SCI |
出版者 | ELECTRONIC JOURNAL OF STATISTICS |
内容类型 | 其他 |
源URL | [http://hdl.handle.net/20.500.11897/475852] |
专题 | 数学科学学院 |
推荐引用方式 GB/T 7714 | Li, Yanfang,Jia, Jinzhu. L1 least squares for sparse high-dimensional LDA. 2017-01-01. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论