Exploring the Limits of Bootstrap

Explores the application of bootstrap to problems that place unusual demands on the method. The bootstrap method, introduced by Bradley Efron in 1973, is a nonparametric technique for inferring the distribution of a statistic derived from a sample. Most of the papers were presented at a special meeting sponsored by the Institute of Mathematical Statistics and the Interface Foundation in May, 1990.

Partial table of contents:

GENERAL PRINCIPLES OF THE BOOTSTRAP.

On the Bootstrap of M-Estimators and Other Statistical Functionals(M. Arcones & E. Gine).

Bootstrapping Markov Chains (K. Athreya & C. Fuh).

Six Questions Raised by the Bootstrap (B. Efron).

Efficient Bootstrap Simulation (P. Hall).

Bootstrapping Signs (R. LePage).

Bootstrap Bandwidth Selection (J. Marron).

APPLICATIONS OF THE BOOTSTRAP.

A Generalized Bootstrap (E. Bedrick & J. Hill).

Bootstrapping Admissible Linear Model Selection Procedures (D. Brownstone).

A Hazard Process for Survival Analysis (J. Hsieh).

A Nonparametric Density Estimation Based Resampling Algorithm (M. Taylor & J. Thompson).

Nonparametric Rank Estimation Using Bootstrap Resampling andCanonical Correlation Analysis (X. Tu, et al.).

Index.