Robust and sparse estimation methods for high-dimensional linear and logistic regression

Creative Commons License

KURNAZ F. S., Hoffmann I., Filzmoser P.

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, vol.172, pp.211-222, 2018 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 172
  • Publication Date: 2018
  • Doi Number: 10.1016/j.chemolab.2017.11.017
  • Journal Name: CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.211-222
  • Keywords: Elastic net penalty, Least trimmed squares, C-step algorithm, High-dimensional data, Robustness, Sparse estimation, LARGE DATA SETS, SQUARES REGRESSION, REGULARIZATION, SELECTION, MODELS
  • Yıldız Technical University Affiliated: Yes

Abstract

Fully robust versions of the elastic net estimator are introduced for linear and logistic regression. The algorithms used to compute the estimators are based on the idea of repeatedly applying the non-robust classical estimators to data subsets only. It is shown how outlier-free subsets can be identified efficiently, and how appropriate tuning parameters for the elastic net penalties can be selected. A final reweighting step improves the efficiency of the estimators. Simulation studies compare with non-robust and other competing robust estimators and reveal the superiority of the newly proposed methods. This is also supported by a reasonable computation time and by good performance in real data examples.