Robust stepwise regression
- 1 August 2002
- journal article
- research article
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 29 (6) , 825-840
- https://doi.org/10.1080/02664760220136168
Abstract
The selection of an appropriate subset of explanatory variables to use in a linear regression model is an important aspect of a statistical analysis. Classical stepwise regression is often used with this aim but it could be invalidated by a few outlying observations. In this paper, we introduce a robust F-test and a robust stepwise regression procedure based on weighted likelihood in order to achieve robustness against the presence of outliers. The introduced methodology is asymptotically equivalent to the classical one when no contamination is present. Some examples and simulation are presented.Keywords
This publication has 13 references indexed in Scilit:
- Robust model selection in regression via weighted likelihood methodologyStatistics & Probability Letters, 2002
- Weighted Likelihood Equations with Bootstrap Root SearchJournal of the American Statistical Association, 1998
- R: A Language for Data Analysis and GraphicsJournal of Computational and Graphical Statistics, 1996
- Efficiency Versus Robustness: The Case for Minimum Hellinger Distance and Related MethodsThe Annals of Statistics, 1994
- Selection of Subsets of Regression VariablesJournal of the Royal Statistical Society. Series A (General), 1984
- Some Comments on C PTechnometrics, 1973
- The discarding of variables in multivariate analysisBiometrika, 1967
- The Best Sub-Set in Multiple Regression AnalysisJournal of the Royal Statistical Society Series C: Applied Statistics, 1965
- Stepwise Least Squares: Residual Analysis and Specification ErrorJournal of the American Statistical Association, 1961
- Note on Stepwise Least SquaresJournal of the American Statistical Association, 1961