The principal problem with principal components regression. Issue 1 (1st January 2019)

• Record Type: Journal Article
• Authors: Artigue, Heidi; Smith, Gary
• Editors: Lu, Zudi
• Is Part Of: Cogent mathematics & statistics. Volume 6:Issue 1(2019)
• Publication Date: 2019-01-01
• Subjects: principal components regression -- PCA -- factor analysis -- Big Data -- data reduction; Mathematics -- Periodicals; Statistics -- Periodicals; Mathematics; Statistics; Periodicals; 510
• DOI: 10.1080/25742558.2019.1622190
• Languages: English
• ISSNs: 2574-2558

Abstract

Abstract: Principal components regression (PCR) reduces a large number of explanatory variables in a regression model down to a small number of principal components. PCR is thought to be more useful, the more numerous the potential explanatory variables. The reality is that a large number of candidate explanatory variables does not make PCR more valuable; instead, it magnifies the failings of PCR.