Robustness to outliers in location-scale parameter model using log-regularly varying distributions

Abstract

Estimating the location and scale parameters is common in statistics, using, for instance, the well-known sample mean and standard deviation. However, inference can be contaminated by the presence of outliers if modeling is done with light-tailed distributions such as the normal distribution. In this paper, we study robustness to outliers in location-scale parameter models using both the Bayesian and frequentist approaches. We find sufficient conditions (e.g., on tail behavior of the model) to obtain whole robustness to outliers, in the sense that the impact of the outliers gradually decreases to nothing as the conflict grows infinitely. To this end, we introduce the family of log-Pareto-tailed symmetric distributions that belongs to the larger family of log-regularly varying distributions.

Published at http://dx.doi.org/10.1214/15-AOS1316 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (10)

• A New Bayesian Approach to Robustness Against Outliers in Linear Regression
• An automatic robust Bayesian approach to principal component regression
• Informed reversible jump algorithms
• Theoretical properties of Bayesian Student-$t$ linear regression
• Robustness against conflicting prior information in regression
• Posterior Robustness with Milder Conditions: Contamination Models Revisited
• Robust heavy-tailed versions of generalized linear models with applications in actuarial science
• Quantile Least Squares: A Flexible Approach for Robust Estimation and Validation of Location-Scale Families
• A Simple Proof of Posterior Robustness
• An Approximate Identity Link Function for Bayesian Generalized Linear Models