A Simple Diagnostic Plot Connecting Robust Estimation, Outlier Detection, and False Discovery Rates
Authors:
- 1 December 2006
- journal article
- research article
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 33 (10) , 1131-1147
Abstract
Robust estimation of parameters, and identification of specific data points that are discordant with an assumed model, are often treated as different statistical problems. The two aims are, however, closely inter-related and in many cases the two analyses are required simultaneously. We present a simple diagnostic plot that connects existing robust estimators with simultaneous outlier detection, and uses the concept of false discovery rates to allow for the multiple comparisons induced by considering each point as a potential outlier. It is straightforward to implement, and applicable in any situation for which robust estimation procedures exist. Several examples are given.Keywords
This publication has 28 references indexed in Scilit:
- Statistical significance for genomewide studiesProceedings of the National Academy of Sciences, 2003
- Outlier Detection and False Discovery Rates for Whole-Genome DNA MatchingJournal of the American Statistical Association, 2003
- A Direct Approach to False Discovery RatesJournal of the Royal Statistical Society Series B: Statistical Methodology, 2002
- The control of the false discovery rate in multiple testing under dependencyThe Annals of Statistics, 2001
- M-Estimates of regression when the scale is unknown and the error distribution is possibly asymmetric: A minimax resultCanadian Journal of Statistics, 1996
- The Maximum Size of Standardized and Internally Studentized Residuals in Regression AnalysisThe American Statistician, 1994
- The Use and Interpretation of Residuals Based on Robust EstimationJournal of the American Statistical Association, 1993
- Procedures for the Identification of Multiple Outliers in Linear ModelsJournal of the American Statistical Association, 1993
- Conditionally Unbiased Bounded-Influence Estimation in General Regression Models, with Applications to Generalized Linear ModelsJournal of the American Statistical Association, 1989
- Robust Estimation of a Location ParameterThe Annals of Mathematical Statistics, 1964