A note on marginal correlation based screening. (10th December 2020)

• Record Type: Journal Article
• Authors: Wang, Run; Dutta, Somak; Roy, Vivekananda
• Is Part Of: Statistical analysis and data mining. Volume 14:Number 1(2021)
• Page Start: 88
• Page End: 92
• Publication Date: 2020-12-10
• Subjects: correlation -- feature selection -- screening -- sure independence screening -- two‐sample t‐test; Data mining -- Statistical methods -- Periodicals; 006.312
• DOI: 10.1002/sam.11491
• Languages: English
• ISSNs: 1932-1864

Abstract

Abstract: Independence screening methods such as the two‐sample t ‐test and the marginal correlation based ranking are among the most widely used techniques for variable selection in ultrahigh‐dimensional data sets. In this short note, simple examples are used to demonstrate potential problems with the independence screening methods in the presence of correlated predictors. Also, an example is considered where all important variables are independent among themselves and all but one important variables are independent with the unimportant variables. Furthermore, a real data example from a genome‐wide association study is used to illustrate inferior performance of marginal correlation screening compared to another screening method.