Jackknife empirical likelihood test for high-dimensional regression coefficients

Yangguang Zang, Sanguo Zhang, Qizhai Li and Qingzhao Zhang

Computational Statistics & Data Analysis, 2016, vol. 94, issue C, 302-316

Abstract: A novel way to test coefficients in high-dimensional linear regression model is presented. Under the ‘large p small n’ situation, the traditional methods, like F-test and t-test, are unsuitable or undefined. The proposed jackknife empirical likelihood test has an asymptotic chi-square distribution and the conditions are much weaker than those in the existing methods. Moreover, an extension of the proposed method can test part of the regression coefficients, which is practical in considering the significance for a subset of covariates. Simulations show that the proposed test has a good control of the type-I error, and is more powerful than Zhong and Chen (2011)’s method in most cases. The proposed test is employed to analyze a rheumatoid arthritis data to find the association between rheumatoid arthritis and the SNPs on the chromosomes 6.

Keywords: Jackknife empirical likelihood; High-dimensional analysis; Regression coefficients; Partial test with nuisance parameter; Power; Type-I error rate (search for similar items in EconPapers)
Date: 2016
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0167947315002005
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:94:y:2016:i:c:p:302-316

DOI: 10.1016/j.csda.2015.08.012

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 ().