Asymptotic nonequivalence of some bandwidth selectors in nonparametric regression
- 1 August 1985
- journal article
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 72 (2) , 481-484
- https://doi.org/10.1093/biomet/72.2.481
Abstract
SUMMARY The bandwidth selection problem in nonparametric kernel regression is considered. Bandwidth selectors based on cross-validation and on Akaike's information criterion, atc, and his finite prediction error, FBE, are among those compared. It is seen that they are not necessarily asymptotically equivalent. Conditions are given under which the equivalence holds and modifications are suggested which make the selectors equivalent.Keywords
This publication has 0 references indexed in Scilit: