Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/31077 
Year of Publication: 
2003
Series/Report no.: 
Discussion Paper No. 359
Publisher: 
Ludwig-Maximilians-Universität München, Sonderforschungsbereich 386 - Statistische Analyse diskreter Strukturen, München
Abstract: 
In recent years the introduction of aggregation methods led to many new techniques within the field of prediction and classification. The most important developments, bagging and boosting, habe been extensively analyzed for two and multi class problems. While the proposed methods treat the class indicator as a nominal response without any structure, in many applications the class may be considered as a ordered categorical variable. In the present paper variants of bagging and boosting are proposed which make use of the ordinal structure. It is demonstrated how the predictive power is improved by use of appropriate aggregation methods. Comparisons between the methods are based on misclassification rates as well as criteria that take ordinality into account, like absolute or squared distance measures.
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size
277.94 kB
302.16 kB





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