Journal of the American Statistical Association · 1978 · 82 citations · 6 references
Natural SciencesBiostatisticsMathematicsBayesian Data AnalysisStatisticsUncertainty QuantificationEstimation TheoryInferential StatisticsRobust Statistical ModelingRobust StatisticsStatistical AnalysisStatistical InferenceCovarianceLinear Discriminant AnalysisMonte Carlo EstimationQuadratic Discriminant AnalysisRobust Covariance EstimationRobust EstimationRobust Linear ModelsRobust Multivariate Analysis
Abstract Two new methods of constructing robust linear and quadratic discriminant functions are introduced. The first is a generalization of Fisher's procedure for finding a linear discriminant function. It places less weight on those observations that are far from the overlapping regions of the two populations. The second new method substitutes M-estimates of the means and the covariance matrices into the usual expressions for the linear and quadratic discriminant functions. Monte Carlo results indicate lower misclassification probabilities for these schemes compared to Fisher's linear discriminant function in cases of heavy-tailed or contaminated distributions.
6
A Rapidly Convergent Descent Method for Minimization
R. Fletcher, M. J. D. Powell · The Computer Journal · 1963 · 4.6K citations · Full text
Discriminant Analysis Based on Ranks
Ronald H. Randles, James D. Broffitt, John S. Ramberg et al. · Journal of the American Statistical Association · 1978 · 46 citations