A new variable selection approach using Random Forests

A. Hapfelmeier and K. Ulm

Computational Statistics & Data Analysis, 2013, vol. 60, issue C, 50-69

Abstract: Random Forests are frequently applied as they achieve a high prediction accuracy and have the ability to identify informative variables. Several approaches for variable selection have been proposed to combine and intensify these qualities. An extensive review of the corresponding literature led to the development of a new approach that is based on the theoretical framework of permutation tests and meets important statistical properties. A comparison to another eight popular variable selection methods in three simulation studies and four real data applications indicated that: the new approach can also be used to control the test-wise and family-wise error rate, provides a higher power to distinguish relevant from irrelevant variables and leads to models which are located among the very best performing ones. In addition, it is equally applicable to regression and classification problems.

Keywords: Random Forests; Variable selection; Permutation tests; Multiple testing (search for similar items in EconPapers)
Date: 2013
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (23)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0167947312003490
Full text for ScienceDirect subscribers only.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:csdana:v:60:y:2013:i:c:p:50-69

DOI: 10.1016/j.csda.2012.09.020

Access Statistics for this article

Computational Statistics & Data Analysis is currently edited by S.P. Azen

More articles in Computational Statistics & Data Analysis from Elsevier
Bibliographic data for series maintained by Catherine Liu ().