A Theoretical Analysis of the Peaking Phenomenon in Classification

Amin Zollanvari (), Alex Pappachen James and Reza Sameni
Additional contact information
Amin Zollanvari: Nazarbayev University
Alex Pappachen James: Nazarbayev University
Reza Sameni: Shiraz University

Journal of Classification, 2020, vol. 37, issue 2, No 10, 434 pages

Abstract: Abstract In this work, we analytically study the peaking phenomenon in the context of linear discriminant analysis in the multivariate Gaussian model under the assumption of a common known covariance matrix. The focus is finite-sample setting where the sample size and observation dimension are comparable. Therefore, in order to study the phenomenon in such a setting, we use an asymptotic technique whereby the number of sample points is kept comparable in magnitude to the dimensionality of observations. The analysis provides a more thorough picture of the phenomenon. In particular, the analysis shows that as long as the Relative Cumulative Efficacy of an additional Feature set (RCEF) is greater (less) than the size of this set, the expected error of the classifier constructed using these additional features will be less (greater) than the expected error of the classifier constructed without them. Our result highlights underlying factors of the peaking phenomenon relative to the classifier used in this study and, at the same time, calls into question the classical wisdom around the peaking phenomenon.

Keywords: Peaking phenomenon; Linear discriminant analysis; Classification error rate; Multiple asymptotic analysis (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s00357-019-09327-3 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

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:spr:jclass:v:37:y:2020:i:2:d:10.1007_s00357-019-09327-3

Ordering information: This journal article can be ordered from
http://www.springer. ... hods/journal/357/PS2

DOI: 10.1007/s00357-019-09327-3

Access Statistics for this article

Journal of Classification is currently edited by Douglas Steinley

More articles in Journal of Classification from Springer, The Classification Society
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().