Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/77188 
Year of Publication: 
1998
Series/Report no.: 
Technical Report No. 1998,22
Publisher: 
Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen, Dortmund
Abstract: 
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.
Subjects: 
Dimension reduction
Outliers
High breakdown procedures
Document Type: 
Working Paper

Files in This Item:
File
Size
1.34 MB
291.4 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.