Open Access
March, 1992 Efficiency and Robustness in Resampling
Regina Y. Liu, Kesar Singh
Ann. Statist. 20(1): 370-384 (March, 1992). DOI: 10.1214/aos/1176348527


It is known that the standard delete-1 jackknife and the classical bootstrap are in general equally efficient for estimating the mean-square-error of a statistic in the i.i.d. setting. However, this equivalence no longer holds in the linear regression model. It turns out that the bootstrap is more efficient when error variables are homogeneous and the jackknife is more robust when they are heterogeneous. In fact, we can divide all the commonly used resampling procedures for linear regression models into two types: the E-type (the efficient ones like the bootstrap) and the R-type (the robust ones like the jackknife). Thus the theory presented here provides a unified view of all the known resampling procedures in linear regression.


Download Citation

Regina Y. Liu. Kesar Singh. "Efficiency and Robustness in Resampling." Ann. Statist. 20 (1) 370 - 384, March, 1992.


Published: March, 1992
First available in Project Euclid: 12 April 2007

zbMATH: 0755.62038
MathSciNet: MR1150349
Digital Object Identifier: 10.1214/aos/1176348527

Primary: 62G15

Keywords: Asymptotic relative efficiency , asymptotic variance , Bootstrap procedures , E-type , jackknife procedures , Resampling , R-type

Rights: Copyright © 1992 Institute of Mathematical Statistics

Vol.20 • No. 1 • March, 1992
Back to Top