Abstract:This paper addresses the problem of robust Gaussian mixture modeling in the presence of outliers. We commence by introducing a general expectation-maximization (EM)-like scheme, called -BM, for iterative numerical computation of the minimum -divergence estimator (M DE). This estimator leverages Parzen's non-parametric ernel density estimate to down-weight low density regions associated with outlying measurements. Akin to the conventional EM, the -BM involves successive Maximizations of lower Bounds on the objective function of the M DE. However, differently from EM, these bounds are not exclusively reliant on conditional expectations. The -BM algorithm is applied to robust parameter estimation of a finite-order multivariate Gaussian mixture model (GMM). We proceed by introducing a new robust variant of the Bayesian information criterion (BIC) that penalizes the M DE's objective function. The proposed criterion, called -BIC, is conveniently applied for robust GMM order selection. In the paper, we also establish a data-driven procedure for selection of the kernel's bandwidth parameter. This procedure operates by minimizing an empirical asymptotic approximation of the mean-integrated-squared-error (MISE) between the underlying density and the estimated GMM density. Lastly, the -BM, the -BIC, and the MISE based selection of the kernel's bandwidth are combined into a unified framework for joint order selection and parameter estimation of a GMM. The advantages of the -divergence based framework over other robust approaches are illustrated in simulation studies involving synthetic and real data.

A robust multivariate Birnbaum–Saunders distribution: EM estimation

Bayesian estimation and prediction for certain mixtures

The multivariate tail-inflated normal distribution and its application in finance

Theoretical results and modeling under the discrete Birnbaum-Saunders distribution

Birnbaum–Saunders frailty regression models for clustered survival data

On Birnbaum-Saunders Inference

Parameter Estimation of Birnbaum-Saunders Distribution under Competing Risks Using the Quantile Variant of the Expectation-Maximization Algorithm

An EM algorithm for estimating the parameters of the multivariate skew-normal distribution with censored responses

t distribution-based robust semiparametric mixture regression model

Statistical inference of the Birnbaum-Saunders model using adaptive progressively hybrid censored data and its applications

Optimal B-Robust Estimation for the Parameters of Marshall-Olkin Extended Burr XII Distribution and Application for Modeling Data from Pharmacokinetics Study

Em Algorithm of the Truncated Multinormal Distribution with Linear Restriction on the Variables

Birnbaum-Saunders parameters estimation using simple random sampling and ranked set sampling

Estimation and Influence Diagnostics for the Multivariate Linear Regression Models with Skew Scale Mixtures of Normal Distributions

BCMA-ES: A Bayesian approach to CMA-ES

Robust Gaussian Mixture Modeling: A -Divergence Based Approach

Birnbaum Saunders distribution for imprecise data: statistical properties, estimation methods, and real life applications

Bimodal Birnbaum–Saunders generalized autoregressive score model

Scale-mixture Birnbaum-Saunders quantile regression models applied to personal accident insurance data

A new class of survival regression models with heavy-tailed errors: robustness and diagnostics

Robust Bayesian Synthetic Likelihood via a Semi-Parametric Approach