Non-convex penalized estimation in high-dimensional models with single-index structure

Tao Wang, Pei Rong Xu, Lixing ZHU*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

24 Citations (Scopus)

Abstract

As promising alternatives to the LASSO, non-convex penalized methods, such as the SCAD and the minimax concave penalty method, produce asymptotically unbiased shrinkage estimates. By adopting non-convex penalties, in this paper we investigate uniformly variable selection and shrinkage estimation for several parametric and semi-parametric models with single-index structure. The new method does not need to estimate the involved nonparametric transformation or link function. The resulting estimators enjoy the oracle property even in the "large p, small n" scenario. The theoretical results for linear models are in parallel extended to general single-index models with no distribution constraint for the error at the cost of mild conditions on the predictors. Simulation studies are carried out to examine the performance of the proposed method and a real data analysis is also presented for illustration.

Original languageEnglish
Pages (from-to)221-235
Number of pages15
JournalJournal of Multivariate Analysis
Volume109
DOIs
Publication statusPublished - Aug 2012

Scopus Subject Areas

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

User-Defined Keywords

  • High-dimensional variable selection
  • Minimax concave penalty
  • Oracle property
  • Penalized least squares
  • SCAD
  • Single-index model

Fingerprint

Dive into the research topics of 'Non-convex penalized estimation in high-dimensional models with single-index structure'. Together they form a unique fingerprint.

Cite this