Bootstrap: A Glance at the Method That Revolutionized Statistics

Mr. Marco Vaassen
University of Wisconsin-Milwaukee
PhD Graduate Student – Teaching Assistant

“Using an impressively simple idea, Bradley Efron, in 1979, gave statisticians a powerful framework for estimating properties of an unknown distribution that he called “the bootstrap”. At the heart of every bootstrapping method is its resampling scheme, the procedure of sampling from a known approximate distribution generated by a sample from an unknown distribution. It is this procedure resembling “pulling yourself up by your bootstraps” that gave the method its name. Owed to its generality, bootstrapping can be applied to simple point estimation, computation of confidence bands, hypothesis testing, and regression.

We will introduce the core ideas of bootstrapping and explore what we can do with it.”