A simulation study of the use of principal components in linear discriminant analysis
- 1 November 1996
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 55 (4) , 353-366
- https://doi.org/10.1080/00949659608811775
Abstract
The use of principal components in linear discriminant analysis has been considered by several authors. However, many of the properties of procedures associated with this usage are unknown. Here, using a simulation study, we investigate the power to choose the ‘correct’ components in the discriminant function. We also include the null case where no component has any discriminatory value.Keywords
This publication has 4 references indexed in Scilit:
- Discriminant Analysis and Statistical Pattern RecognitionWiley Series in Probability and Statistics, 1992
- On the Use of Component Scores in the Presence of Group StructureJournal of Consumer Research, 1989
- On Using Principal Components Before Separating a Mixture of Two Multivariate Normal DistributionsJournal of the Royal Statistical Society Series C: Applied Statistics, 1983
- Between-group comparison of principal components — some sampling resultsJournal of Statistical Computation and Simulation, 1982