期刊
STATISTICS & PROBABILITY LETTERS
卷 184, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.spl.2022.109379
关键词
Precision matrix; High-dimensional compositional data; Centered log-ratio transformation; Sparsity; Huber robustness
资金
- National Natural Science Foundation of China [11701294, 62061040]
- Fundamental Research Funds for the Central Universities of China [63206022]
- Candidate Research Innovation Fund of Nankai University
Motivated by the rapid development in high-dimensional compositional data analysis, this paper proposes an Approximate-Plug framework with theoretical justifications for robust precision matrix estimation under the sparsity assumption. The proposed method constructs a robust estimator and utilizes a constrained minimization procedure to obtain the final estimator. Simulation studies and real data application demonstrate the superiority of the proposed method over existing approaches.
Motivated by the rapid development in the high-dimensional compositional data analysis, an Approximate-Plug framework with theoretical justifications is proposed to provide robust precision matrix estimation for this kind of data under the sparsity assumption. To be specific, we first construct a Huber-robustness estimator ((gamma) over tilde)& nbsp;to approximate the centered log-ratio covariance matrix. Then we plug ((gamma) over tilde) into a constrained l1-minimization procedure to obtain the final estimator tilde ((omega) over tilde). Through imposing some mild conditions, we derive the convergence rate under the entrywise maximum norm and the spectral norm. Given that SpiecEasi in Kurtz et al. (2015) shares same routine with us but lacks of robustness and theoretical guarantees, simulation studies are conducted to show the privileges of our procedure. We also apply the proposed method on a real data. (c) 2022 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据