Solving classification problems with the partial least squares method and classification trees /

Bibliographic Details
Main Author:	Yeh, Chyon-Hwa, 1962-
Other Authors:	Carroll, Raymond J. (degree committee member.), Hsing, Tailen (degree committee member.), Tretter, Marietta J. (degree committee member.)
Format:	Thesis Book
Language:	English
Published:	1991.
Subjects:	Discriminant analysis. Trees (Graph theory) Least squares. Major statistics. Academic theses
Online Access:	Link to OAKTrust copy

Description
Abstract:	High dimensional data are found in scientific fields. Difficulties arise when one applies classical classification methods to these high dimensional data sets, because of multicollinearity. Problems with high dimensional data sets can be overcome by reducing the dimensions of data sets. The partial least squares method (PLS) is a new method used for dimension reduction. The statistical properties and limiting distribution of some PLS components are derived. The classification and regression trees method (CART) is applied to the reduced data for solving classification problems. A new stopping criterion for the PLS procedure is introduced.
Item Description:	Typescript (photocopy). Vita. "Major subject: Statistics."
Physical Description:	viii, 75 leaves : illustrations ; 29 cm
Bibliography:	Includes bibliographical references.