Robust sliced inverse regression procedures

Ursula Gather; Torsten Hilker; Claudia Becker

Sliced Inverse Regression (SIR) is a promising technique for the purpose of dimension reduction. Several properties of this relatively new method have been examined already, but little attention has been paid to robustness aspects. We show that SIR is very sensitive towards outliers in the data. Therefore a generalized estimation procedure which allows for robustness properties, especially for a high breakdown point, is proposed.

MoreLess

Year of publication:	1998
Authors:	Gather, Ursula ; Hilker, Torsten ; Becker, Claudia
Publisher:	Dortmund : SFB 475, Universität Dortmund
Subject:	Dimension reduction \| Outliers \| High breakdown procedures \| Schätztheorie \| Estimation theory \| Regressionsanalyse \| Regression analysis \| Robustes Verfahren \| Robust statistics