Outlier detection and robust estimation of scale
- 1 January 1987
- journal article
- research article
- Published by Informa UK Limited in Journal of Statistical Computation and Simulation
- Vol. 27 (1), 79-92
- https://doi.org/10.1080/00949658708810981
Abstract
The strategy of removing points via a backwards-stepping outlier detection procedure and then taking the standard deviation of the remaining points is examined as a robust scale estimator via computer simulations. It is shown that this procedure compares favorably with the most effective robust scale estimators. This is particularly true if the outliers are relatively extreme or follow an asymmetric distribution. It is also shown that this strategy results in an estimator with high breakdown and redescending influence.Keywords
This publication has 9 references indexed in Scilit:
- Robust Estimators of Scale: Finite-Sample Performance in Long-Tailed Symmetric DistributionsJournal of the American Statistical Association, 1985
- Finite Sample Breakdown of $M$- and $P$-EstimatorsThe Annals of Statistics, 1984
- A comparison of robust methods and detection of outliers techniques when estimating a location parameterCommunications in Statistics - Theory and Methods, 1984
- Robustness properties for a class of scale estimatorsCommunications in Statistics - Theory and Methods, 1984
- The calculation of outlier detection statisticsCommunications in Statistics - Simulation and Computation, 1984
- Robust estimates and tests for the one- and two-sample scale modelsBiometrika, 1982
- Robust StatisticsWiley Series in Probability and Statistics, 1981
- Computer Simulation Swindles, with Applications to Estimates of Location and DispersionJournal of the Royal Statistical Society Series C: Applied Statistics, 1976
- On the Detection of Many OutliersTechnometrics, 1975