Outlier detection by robust principal components analysis
- 1 January 2000
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 29 (1) , 139-151
- https://doi.org/10.1080/03610910008813606
Abstract
The robust principal components analysis (RPCA) introduced by Campbell (Applied Statistics 1980, 29, 231–237) provides in addition to robust versions of the usual output of a principal components analysis, weights for the contribution of each point to the robust estimation of each component. Low weights may thus be used to indicate outliers. The present simulation study provides critical values for testing the kth smallest weight in the RPCA of a sample of n p-dimensional vectors, under the null hypothesis of a multivariate normal distribution. The cases p=2(2)10, 15, 20 for n=20, 30, 40, 50, 75, 100 subject to n≥p/2, are examined, with k≤√n.Keywords
This publication has 11 references indexed in Scilit:
- Union-Intersection testing for outliers in multivariate normal dataJournal of Statistical Computation and Simulation, 1995
- Identifying Multiple Outliers in Multivariate DataJournal of the Royal Statistical Society Series B: Statistical Methodology, 1992
- Sequential Application of Wilks's Multivariate Outlier TestJournal of the Royal Statistical Society Series C: Applied Statistics, 1992
- Multivariate outlier tests with structured co variance matricesJournal of Statistical Computation and Simulation, 1991
- Unmasking Multivariate Outliers and Leverage PointsJournal of the American Statistical Association, 1990
- A New Graphical Method for Detecting Single and Multiple Outliers in Univariate and Multivariate DataJournal of the Royal Statistical Society Series C: Applied Statistics, 1987
- Robust Methods in the Assessment of Multivariate NormalityJournal of the Royal Statistical Society Series C: Applied Statistics, 1984
- Some Techniques for Assessing Multivarate Normality Based on the Shapiro- Wilk WJournal of the Royal Statistical Society Series C: Applied Statistics, 1983
- Identification of OutliersPublished by Springer Nature ,1980
- Robust Procedures in Multivariate Analysis I: Robust Covariance EstimationJournal of the Royal Statistical Society Series C: Applied Statistics, 1980