English

Penalized Composite Quasi-Likelihood for Ultrahigh-Dimensional Variable Selection

Methodology 2011-07-06 v2 Statistics Theory Statistics Theory

Abstract

In high-dimensional model selection problems, penalized simple least-square approaches have been extensively used. This paper addresses the question of both robustness and efficiency of penalized model selection methods, and proposes a data-driven weighted linear combination of convex loss functions, together with weighted L1L_1-penalty. It is completely data-adaptive and does not require prior knowledge of the error distribution. The weighted L1L_1-penalty is used both to ensure the convexity of the penalty term and to ameliorate the bias caused by the L1L_1-penalty. In the setting with dimensionality much larger than the sample size, we establish a strong oracle property of the proposed method that possesses both the model selection consistency and estimation efficiency for the true non-zero coefficients. As specific examples, we introduce a robust method of composite L1-L2, and optimal composite quantile method and evaluate their performance in both simulated and real data examples.

Keywords

Cite

@article{arxiv.0912.5200,
  title  = {Penalized Composite Quasi-Likelihood for Ultrahigh-Dimensional Variable Selection},
  author = {Jelena Bradic and Jianqing Fan and Weiwei Wang},
  journal= {arXiv preprint arXiv:0912.5200},
  year   = {2011}
}
R2 v1 2026-06-21T14:28:53.073Z