A0298
Title: When does massive data bootstrap work
Authors: Patrice Bertail - University of Paris-Ouest-Nanterre-La Defense (France)
Liuhua Peng - The University of Melbourne (Australia)
Dimitris Politis - University of California, San Diego (USA)
Han Lin Shang - Macquarie University (Australia)
Stanislav Volgushev - University of Toronto (Canada)
Nan Zou - Macquarie University (Australia) [presenting]
Abstract: In classic statistical inference, the bootstrap stands out as a simple, powerful, and data-driven technique. However, when coping with massive data sets, which are increasingly prevalent these days, the bootstrap can be computationally infeasible. To speed up the bootstrap for massive data sets, the bag of little bootstraps was invented in 2014. Despite its considerable popularity, little is known about the theoretical properties of the bag of little bootstrap, including reliability. Indeed, the preliminary results have already raised questions on the applicability of the bag of little bootstraps under a simple but important setting. The procedure for the bag of little bootstraps is first introduced, and then its theoretical applicability is investigated. Specifically, for this applicability, a counterexample for the claimed sufficient condition is presented in the literature and, as a remedy, a hopefully correct, generic sufficient condition is provided.
