期刊论文

Yuanyuan Lin

英文

Journal of Business & Economic Statistics

0735-0015

2023

41

3

957-967

10.1080/07350015.2022.2097911

National Natural Science Foundation of China [12001219, 12171463, 12101330, 11961028]; Fundamental Research Funds for the Central Universities, Nankai University [9920200110]; U.S. National Science Foundation [DMS-1916199]; Hong Kong Research Grants Council [14306219, 14306620]; Chinese University of Hong Kong; NIH [ULI TR002345]

本校为其他机构

In this article, we propose a robust signal recovery method for high-dimensional linear log-contrast models, when the error distribution could be heavy-tailed and asymmetric. The proposed method is built on the Huber loss with ℓ1 ℓ1 penalization. We establish the ℓ1 ℓ1 and ℓ2 ℓ2 consistency for the resulting estimator. Under conditions analogous to the irrepresentability condition and the minimum signal strength condition, we prove that the signed support of the slope parameter vector can be recovered with high probability. The finite-sam...