This book introduces the bootstrap method, a statistical technique for estimating the sampling distribution of a statistic, to students with limited advanced mathematics backgrounds. It provides comprehensive coverage of the bootstrap's advantages and disadvantages, including discussions on improved confidence set estimation, error rate estimation in discriminant analysis, and applications to hypothesis testing and estimation problems. The text also highlights the limitations of the bootstrap through counterexamples and draws connections between traditional resampling methods and the bootstrap, often with accompanying R routines. The primary focus is on using the bootstrap as an exploratory tool, particularly for variable selection and modeling.

The book is designed to offer both an elementary and a modern introduction to bootstrap methods, catering to students without extensive advanced mathematics backgrounds. It aims to provide reliable, hands-on coverage of the method's benefits and drawbacks, distinguishing itself by presenting results on improved confidence set estimation and applications in discriminant analysis and hypothesis testing. The text also carefully addresses the limitations of the bootstrap by exhibiting counterexamples to its consistency, ensuring a balanced understanding for the reader.