A Comparison of Depth Functions in Maximal Depth Classification Rules

Authors

Olusola Samuel Makinde, Federal University of Technology, Akure, NigeriaFollow
Adeyinka Damilare Adewumi, Federal University of Technology Akure, NigeriaFollow

Abstract

Data depth has been described as alternative to some parametric approaches in analyzing many multivariate data. Many depth functions have emerged over two decades and studied in literature. In this study, a nonparametric approach to classification based on notions of different data depth functions is considered and some properties of these methods are studied. The performance of different depth functions in maximal depth classifiers is investigated using simulation and real data with application to agricultural industry.

DOI

10.22237/jmasm/1493598120

Recommended Citation

Makinde, O. S. & Adewumi, A. D. (2017). A comparison of depth functions in maximal depth classification rules. Journal of Modern Applied Statistical Methods, 16(1), 388-405. doi: 10.22237/jmasm/1493598120