Robust Model-Based Clustering

TL;DR

This paper introduces a new class of robust, Fisher-consistent estimators for mixture models, enhancing model-based clustering with improved robustness and accuracy, especially for multivariate normal mixtures.

Contribution

It develops a novel robust estimation method for mixture models, including an algorithm similar to EM, and demonstrates superior performance through simulations and real data applications.

Findings

Our estimators outperform existing methods in robustness and accuracy.

The proposed algorithm effectively computes clusters in multivariate normal mixtures.

Real data analysis confirms the method's practical advantages.

Abstract

We propose a new class of robust and Fisher-consistent estimators for mixture models. These estimators can be used to construct robust model-based clustering procedures. We study in detail the case of multivariate normal mixtures and propose a procedure that uses S estimators of multivariate location and scatter. We develop an algorithm to compute the estimators and to build the clusters which is quite similar to the EM algorithm. An extensive Monte Carlo simulation study shows that our proposal compares favorably with other robust and non robust model-based clustering procedures. We apply ours and alternative procedures to a real data set and again find that the best results are obtained using our proposal.