Small sample properties of bootstrap

Abstract

The Bootstrap 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 'he parameters and their 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.