Events and Meetings of Italian Statistical Society, Statistics and Demography: the Legacy of Corrado Gini

Font Size: 
Quantile-based classifiers
Cinzia Viroli

Last modified: 2015-09-05

Abstract


Quantile classifiers are defined as a sum of appropriately weighted component -wise distances to the within-class quantiles. The classifier is particularly suitable for potentially high-dimensional and skewed data. An optimal percentage for the quantiles can be chosen by minimizing the misclassification error in the training sample. It is shown that it in the univariate case it is the optimal decision boundary point that minimizes the Gini transvariation probability.

Full Text: PDF