Partitioned cross-validation
Open Access
- 1 January 1987
- journal article
- research article
- Published by Taylor & Francis in Econometric Reviews
- Vol. 6 (2) , 271-283
- https://doi.org/10.1080/07474938708800136
Abstract
Partitioned cross-validation is proposed as a method for overcoming the large amounts of across sample variability to which ordinary cross-validation is subject. The price for cutting down on the sample noise is that a type of bias is intriduced. A theory is presented for optimal trade-off of this variance and bias. Comparison with other bandwidth selection methods is given.Keywords
This publication has 10 references indexed in Scilit:
- Kernel estimators of regression functionsPublished by Cambridge University Press (CUP) ,2008
- Biased and Unbiased Cross-Validation in Density EstimationJournal of the American Statistical Association, 1987
- Estimation of integrated squared density derivativesStatistics & Probability Letters, 1987
- Extent to which least-squares cross-validation minimises integrated square error in nonparametric density estimationProbability Theory and Related Fields, 1987
- On the Amount of Noise Inherent in Bandwidth Selection for a Kernel Density EstimatorThe Annals of Statistics, 1987
- An improved data-based algorithm for choosing the window width when estimating the density at a pointComputational Statistics & Data Analysis, 1986
- An Asymptotically Optimal Window Selection Rule for Kernel Density EstimatesThe Annals of Statistics, 1984
- Large Sample Optimality of Least Squares Cross-Validation in Density EstimationThe Annals of Statistics, 1983
- Curve EstimatesThe Annals of Mathematical Statistics, 1971
- Density Estimation for Statistics and Data AnalysisPublished by Springer Nature ,1400