Robust Model-Based Clustering

Authors

  • Juan Domingo Gonzalez Acoustic Propagation Department, UNIDEF-CONICET
  • Ricardo A. Maronna Universidad de Buenos Aires
  • Victor J. Yohai Universidad de Buenos Aires
  • Ruben H. Zamar University of British Columbia

DOI:

https://doi.org/10.52933/jdssv.v2i6.47

Keywords:

mixture models, EM–algorithm, scatter S–estimators

Abstract

We propose a class of Fisher-consistent robust estimators for mixture models. These estimators are then used to build a robust model-based clustering procedure. We study in detail the case of multivariate Gaussian mixtures and propose an algorithm, similar to the EM algorithm, to compute the proposed estimators and build the robust clusters. An extensive Monte Carlo simulation study shows that our proposal outperforms other robust and non robust, state of the art, model-based clustering procedures. We apply our proposal to a real data set and show that again it outperforms alternative procedures.

Downloads

Published

2022-10-06

How to Cite

Gonzalez, J. D., Maronna, R., Yohai, V., & Zamar, R. (2022). Robust Model-Based Clustering. Journal of Data Science, Statistics, and Visualisation, 2(6). https://doi.org/10.52933/jdssv.v2i6.47