Performances of some high dimensional regression methods


KURNAZ F. S.

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, vol.50, no.6, pp.1820-1836, 2021 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 50 Issue: 6
  • Publication Date: 2021
  • Doi Number: 10.1080/03610918.2021.1881115
  • Journal Name: COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Business Source Elite, Business Source Premier, CAB Abstracts, Compendex, Computer & Applied Sciences, Veterinary Science Database, zbMATH, Civil Engineering Abstracts
  • Page Numbers: pp.1820-1836
  • Keywords: lasso, L, (1) penalty, Robustness, Variable selection
  • Yıldız Technical University Affiliated: Yes

Abstract

Variable selection is one of the important practical issues for many scientific areas, particularly in chemometrics, where data sets include several hundreds of variables and low number of observations. The aim of this paper is to compare some newly proposed variable selection methods by means of extensive simulation studies and to give some practical hints for use of the compared methods. Furthermore, we underpin the performances of compared methods based on real data examples.