When Does Bootstrap Work?: Asymptotic Results and Simulations

Bootstrap methods are procedures for estimating or approximating the distribution of a statistic based on ideas from resampling and simulation methods. This volume is concerned with the asymptotic behaviour of the bootstrap and investigates the conditions under which the bootstrap works satisfactorily. In particular, the author considers the application of the bootstrap to the estimation of smooth functionals, non-parametric curve estimation, and to linear models. Readers are assumed to have a working familiarity with the basics of bootstrap methods.