DMCA
Graphical tools for quadratic discriminant analysis (2007)
| Venue: | Technometrics |
| Citations: | 4 - 1 self |
BibTeX
@ARTICLE{Pardoe07graphicaltools,
author = {Iain Pardoe and Xiangrong Yin and R. Dennis Cook},
title = {Graphical tools for quadratic discriminant analysis},
journal = {Technometrics},
year = {2007},
pages = {172--183}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Sufficient dimension reduction methods provide effective ways to visualize discriminant anal-ysis problems. For example, Cook and Yin (2001) showed that the dimension reduction method of sliced average variance estimation (save) identifies variates that are equivalent to a quadratic discriminant analysis (qda) solution. This article makes this connection explicit to motivate the use of save variates in exploratory graphics for discriminant analysis. Classification can then be based on the save variates using a suitable distance measure. If the chosen measure is Mahalanobis distance, then classification is identical to qda using the original variables. Just as canonical variates provide a useful way to visualize linear discriminant analysis (lda), so save variates help to visualize qda—this would appear to be particularly useful given the lack of graphical tools for qda in current software. Furthermore, while lda and qda can be sensitive to nonnormality, save is more robust. Key words: canonical variates, classification, dimension reduction, linear discriminant analysis (lda), quadratic discriminant analysis (qda), sliced average variance estima-
Keyphrases
quadratic discriminant analysis graphical tool save variate linear discriminant analysis canonical variate discriminant analysis key word suitable distance measure effective way chosen measure mahalanobis distance current software discriminant anal-ysis problem dimension reduction method sufficient dimension reduction method useful way exploratory graphic sliced average variance estimation dimension reduction original variable sliced average variance estima
