Two Iterative Proximal-Point Algorithms for the Calculus of Divergence-based Estimators with Application to Mixture Models

TL;DR

This paper introduces two proximal-point algorithms based on EM to minimize divergence criteria, resulting in robust estimators for mixture models that outperform traditional EM in the presence of outliers.

Contribution

It proposes novel proximal-point algorithms that relax identifiability conditions and enhance robustness of estimators in mixture models.

Findings

Algorithms converge on Weibull and Gaussian mixtures.

Robust estimators outperform EM with outliers.

Simulations confirm effectiveness across divergences.

Abstract

Estimators derived from an EM algorithm are not robust since they are based on the maximization of the likelihood function. We propose a proximal-point algorithm based on the EM algorithm which aim to minimize a divergence criterion. Resulting estimators are generally robust against outliers and misspecification. An EM-type proximal-point algorithm is also introduced in order to produce robust estimators for mixture models. Convergence properties of the two algorithms are treated. We relax an identifiability condition imposed on the proximal term in the literature; a condition which is generally not fulfilled by mixture models. The convergence of the introduced algorithms is discussed on a two-component Weibull mixture and a two-component Gaussian mixture entailing a condition on the initialization of the EM algorithm in order for the later to converge. Simulations on mixture models using different statistical divergences are provided to confirm the validity of our work and the robustness of the resulting estimators against outliers in comparison to the EM algorithm.