Cover Image

Estimation of the conditional risk in classification: The swapping method

The bias of the empirical error rate in supervised classification is studied. It is shown that this bias can be understood as a covariance between the classification rule and the labeling of the training data. From this result, a new penalized criterion is proposed to perform model selection in classification. Applications of the resulting algorithm to simulated and real data are presented.
Year of publication:
2008
Authors: Daudin, Jean-Jacques ; Mary-Huard, Tristan
Published in:
Computational Statistics & Data Analysis. - Elsevier, ISSN 0167-9473. - Vol. 52.2008, 6, p. 3220-3232
Publisher: Elsevier
Type of publication: Article
Source:
Persistent link: https://www.econbiz.de/10005081907
Similar items by person