Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/22203
Authors: 
Year of Publication: 
2004
Series/Report no.: 
Papers No. 2004,30
Publisher: 
Humboldt-Universität zu Berlin, Center for Applied Statistics and Economics (CASE), Berlin
Abstract: 
Tree-based methods have become one of the most flexible, intuitive, and powerful data analytic tools for exploring complex data structures. The applicationsof these methods are far reaching. They include financial firms (credit cards: Altman, 2002; Frydman et al., 2002, and investments: Pace, 1995; Brennan et al., 2001), manufacturing and marketing companies (Levin et al., 1995), and pharmaceutical companies. The best documented, and arguably most popular uses of tree-based methods are in biomedical research for which classification is a central issue. For example, a clinician or health scientist may be very interested in the following question (Goldman et al., 1996, 1982; Zhang et al., 2001): Is this patient with chest pain suffering a heart attack, or does he simply have a strained muscle? To answer this question, information on this patient must be collected, and a good diagnostic test utilizing such information must be in place. Tree-based methods provide one solution for constructing the diagnostic test.
Document Type: 
Working Paper

Files in This Item:
File
Size
227.98 kB





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