A survey of cross-validation procedures for model selection
Top Cited Papers
Open Access
- 1 January 2010
- journal article
- research article
- Published by Institute of Mathematical Statistics in Statistics Surveys
- Vol. 4 (none), 40-79
- https://doi.org/10.1214/09-ss054
Abstract
Used to estimate the risk of an estimator or to perform model selection, cross-validation is a widespread strategy because of its simplicity and its (apparent) universality. Many results exist on model selection performances of cross-validation procedures. This survey intends to relate these results to the most recent advances of model selection theory, with a particular emphasis on distinguishing empirical statements from rigorous theoretical results. As a conclusion, guidelines are provided for choosing the best cross-validation procedure according to the particular features of the problem in hand.Keywords
This publication has 100 references indexed in Scilit:
- Minimal Penalties for Gaussian Model SelectionProbability Theory and Related Fields, 2006
- Asymptotic Optimality of Likelihood-Based Cross-ValidationStatistical Applications in Genetics and Molecular Biology, 2004
- Bandwidth selection in robust smoothingJournal of Nonparametric Statistics, 1993
- Smoothed cross-validationProbability Theory and Related Fields, 1992
- Data-Driven Bandwidth Choice for Density Estimation Based on Dependent DataThe Annals of Statistics, 1990
- A comparative study of ordinary cross-validation, v-fold cross-validation and the repeated learning-testing methodsBiometrika, 1989
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978
- The Predictive Sample Reuse Method with ApplicationsJournal of the American Statistical Association, 1975
- Some Comments onCpTechnometrics, 1973
- Estimation of Error Rates in Discriminant AnalysisTechnometrics, 1968