Robust Model-Based Clustering

JUAN D. GONZALEZ; RICARDO A. MARONNA; VICTOR J.YOHAI; RUBEN H. ZAMAR

Journal of Data Science, Statistics, and Visualisation

International Association for Statistical Computing

Lugar: La Haya; Año: 2022 vol. 2 p. 1 - 29

2773-0689

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, similarto the EM algorithm, to compute the proposed estimators and build the robust clusters. An extensive Monte Carlo simulation study shows that our proposal outperformsother robust and non-robust, state of the art, model-based clustering procedures. Weapply our proposal to a real data set and show that again it outperforms alternativeprocedure