Small Sample Properties of Bootstrap
Abstract
The Boostrap method is a nonparametric statistical technique for estimating the sampling distribution of estimators of unknown parameters. While the asymptotic theory for bootstrap is well established, this thesis investigates the behavior of the bootstrap for small sample sizes. For the exponential distribution and for normal linear regression the bootstrap estimates of the parameters and variances are compared with the theoretical sampling distributions. The small sample properties of bootstrap confidence intervals using the percentile method and the bias-corrected percentile method are also investigated. Keywords: Exponential distributions, Computerized simulation, Fortran, Subroutines.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1988
- Accession Number
- ADA202187
Entities
People
- Stefan Bernhardt
Organizations
- Naval Postgraduate School