Multiclass linear discriminant analysis with ultrahigh‐dimensional features. Issue 4 (18th June 2019)

• Record Type: Journal Article
• Authors: Li, Yanming; Hong, Hyokyoung G.; Li, Yi
• Is Part Of: Biometrics. Volume 75:Issue 4(2019)
• Page Start: 1086
• Page End: 1097
• Publication Date: 2019-06-18
• Subjects: Fisher's multiclass discriminant analysis -- jointly informative features -- marginally informative features -- multivariate screening -- ultrahigh‐dimensional classification; Biometry -- Periodicals; 570.15195
• DOI: 10.1111/biom.13065
• Languages: English
• ISSNs: 0006-341X

Abstract

Abstract: Within the framework of Fisher's discriminant analysis, we propose a multiclass classification method which embeds variable screening for ultrahigh‐dimensional predictors. Leveraging interfeature correlations, we show that the proposed linear classifier recovers informative features with probability tending to one and can asymptotically achieve a zero misclassification rate. We evaluate the finite sample performance of the method via extensive simulations and use this method to classify posttransplantation rejection types based on patients' gene expressions.