DMCA
Saddlepoint Approximations for Bootstrap Distributions (1998)
| Venue: | Computer Science and Statistics |
| Citations: | 1 - 1 self |
BibTeX
@ARTICLE{Davison98saddlepointapproximations,
author = {Prof A. C. Davison and Diego Kuonen and Diego Kuonen},
title = {Saddlepoint Approximations for Bootstrap Distributions},
journal = {Computer Science and Statistics},
year = {1998},
volume = {28},
pages = {248--253}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We develop fast saddlepoint approximations to bootstrap distributions that work in the presence of an outlier, using a saddlepoint mixture approximation with a rule for using these modified approximations for the average. We look at robust M-estimates of location, before generalizing the presented technique to deal with studentized statistics, e.g. Huber's Proposal 2. Simulations show that our approach outperforms the `brute-force' bootstrap by a significant factor.
