Scientific article
OA Policy
English

High Breakdown Inference in the Mixed Linear Model

Published in

Journal of the American Statistical Association, vol. 101, no. 473, p. 292-300

Publication date

2006

Abstract

Mixed linear models are used to analyze data in many settings. These models have a multivariate normal formulation in most cases. The maximum likelihood estimator (MLE) or the residual MLE (REML) is usually chosen to estimate the parameters. However, the latter are based on the strong assumption of exact multivariate normality. Welsh and Richardson have shown that these estimators are not robust to small deviations from the multivariate normality. This means that in practice a small proportion of data (even only one) can drive the value of the estimates on their own. Because the model is multivariate, we propose a high-breakdown robust estimator for very general mixed linear models that include, for example, covariates. This robust estimator belongs to the class of S-estimators, from which we can derive the asymptotic properties for inference. We also use it as a diagnostic tool to detect outlying subjects. We discuss the advantages of this estimator compared with other robust estimators proposed previously and illustrate its performance with simulation studies and analysis of three datasets. We also consider robust inference for multivariate hypotheses as an alternative to the classical F-test by using a robust score-type test statistic proposed by Heritier and Ronchetti, and study its properties through simulations and analysis of real data.

Keywords

  • Constrained covariance
  • Robust F-test
  • S-estimator
  • Variance components

Citation (ISO format)

COPT, Samuel, VICTORIA-FESER, Maria-Pia. High Breakdown Inference in the Mixed Linear Model. In: Journal of the American Statistical Association, 2006, vol. 101, n° 473, p. 292–300. doi: 10.1198/016214505000000772

Main files (1)

Article (Accepted version)

Identifiers

1079views
1625downloads

Technical informations

Creation

30/04/2010 12:34:00

First validation

30/04/2010 12:34:00

Update

14/03/2023 15:28:36

Status update

14/03/2023 15:28:36

Last indexation

29/10/2024 14:43:03
All rights reserved by Archive ouverte UNIGE and the University of Geneva
OpenAIRE Indexed & Validated