×

Bootstrap and cross-validation estimates of the prediction error for linear regression models. (English) Zbl 0557.62039

The authors provide additional insight into the behaviour of bootstrap estimators of the prediction error for linear regression models with normally distributed observations. A main result is that the bias corrected bootstrap estimator is best unbiased and should be the first choice.
Reviewer: S.Wang

MSC:

62G05 Nonparametric estimation
62J05 Linear regression; mixed models

Software:

alr3
PDFBibTeX XMLCite
Full Text: DOI