Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10524548 | Journal of Multivariate Analysis | 2005 | 20 Pages |
Abstract
The effects of moderate levels of serial correlation on one-sided and ordinary cross-validation in the context of local linear and kernel smoothing is investigated. It is shown both theoretically and by simulation that one-sided cross-validation is much less adversely affected by correlation than is ordinary cross-validation. The former method is a reliable means of window width selection in the presence of moderate levels of serial correlation, while the latter is not. It is also shown that ordinary cross-validation is less robust to correlation when applied to Gasser-Müller kernel estimators than to local linear ones.
Related Topics
Physical Sciences and Engineering
Mathematics
Numerical Analysis
Authors
Jeffrey D. Hart, Cherng-Luen Lee,