MetaCartSign in to MyCiteSeer

Include Citations | Advanced Search | Help

Include Citations | Advanced Search | Help

  Flexible Discriminant Analysis by Optimal Scoring (1994) [72 citations — 11 self]

Download:
Download as a PDF | Download as a PS
by Trevor Hastie, Robert Tibshirani, Andreas Buja
Journal of the American Statistical Association
http://utstat.toronto.edu/reports/tibs/fda.ps
Add To MetaCart

Abstract:

c flAT&T Bell Laboratories Fisher's linear discriminant analysis is a valuable tool for multigroup classification. With a large number of predictors, one can find a reduced number of discriminant coordinate functions that are "optimal " for separating the groups. With two such functions one can produce a classification map that partitions the reduced space into regions that are identified with group membership, and the decision boundaries are linear. This paper is about richer nonlinear classification schemes. Linear discriminant analysis is equivalent to multi-response linear regression using optimal scorings to represent the groups. We obtain nonparametric versions of discriminant analysis by replacing linear regression by any nonparametric regression method. In this way, any multi-response regression technique (such as MARS or neural networks) can be post-processed to improve their classification performence.

Citations

2438 Classification and Regression Trees – Breiman, Friedman, et al. - 1984
635 Generalized Additive Models – Hastie, Tibshirani - 1990
392 Multivariate Analysis – Mardia, Kent, et al. - 1979
273 Connectionist learning procedures – Hinton - 1989
267 Projection Pursuit Regression – Friedman, Stuetzle - 1981
168 Exploratory Projection Pursuit – Friedman - 1987
157 Multivariate adaptive regression splines (with discussion). The Annals of Statistics – Friedman - 1991
149 Classi cation and Regression Trees – Breiman, Friedman, et al. - 1984
149 Analysis of hidden units in a layered network trained to classify sonar targets – Gorman, Sejnowski - 1988
104 Neural networks and related methods for classification – Ripley - 1994
98 Estimating optimal transformations for multiple regression and correlation – Breiman, Friedman - 1985
87 Training stochastic model recognition algorithms as networks can lead to maximum mutual information estimation of parameters – Bridle - 1990
68 Hinging hyperplanes for regression, classification, and function approximation – Breiman - 1993
67 Dynamic Error Propagation Networks – Robinson - 1989
54 Nonlinear multivariate analysis – Gifi - 1990
37 Linear smoothers and additive models (with discussion – Buja, Hastie, et al. - 1989
32 Tree-Structured Classification Via Generalized Discriminant Analysis – Loh, Vanichsetakul - 1988
26 The efficiency of logistic regression compared to normal discriminant analysis – Efron - 1975
19 The method for estimating multivariate functions from noisy data{(with discussion – Breiman - 1991
15 Transforming neural-net output levels to probability distributions – Denker, LeCun - 1991
6 Nonlinear discriminant analysis via scaling and ACE – Breiman, Ihaka - 1984
6 Nonlinear Multivariate Analysis – Gi - 1990
6 Pattern classification using neural networks – Lippman - 1989
5 Discussion of "Flexible parsimonious smoothing and additive modelling" by Friedman and Silverman – Hastie - 1989
3 Discussion of \Flexible parsimonious smoothing and additive modelling" by Friedman and – Hastie - 1989
3 Tree structured classi via generalized discriminant analysis – Loh, Vanichsetakul - 1988
2 Hinging hyperplanes for regression, classi and function approximation – Breiman - 1991
2 Discussion of "Multivariate Adaptive Regression Splines" by Friedman – Owen - 1991
2 Nonlinear multivariate analysis, unpublished manuscript – Gifi - 1981
1 Statistical learning networks: a unifying view. Computer Sci and Statist: Proc of the 21st interface – Barron, Barron - 1988
1 The eciency of multinmial logistic regression compared with multigroup normal discriminant analysis – Bull, Donner - 1987
1 Nonlinear multivariate analysis. unpublished manuscript – Gi - 1981
1 Minimizing GCV/GML scores with mulitple smoothing parameters via newton's method – Gu, Wahba - 1984
1 Discussion of \Projection pursuit" by Huber. Annals of Statistics – Hastie, Tibshirani - 1985
1 Pattern classi using neural networks – Lippman - 1989
1 Discussion of "Projection pursuit" by Huber – Hastie, Tibshirani - 1985