Bayesian bootstraps for massive data

In this article, we present data-subsetting algorithms that allow for the approximate and scalable implementation of the Bayesian bootstrap. They are analogous to two existing algorithms in the frequentist literature: the bag of little bootstraps (Kleiner et al., 2014) and the subsampled double boot...

Descripción completa

Detalles Bibliográficos
Autores: Barrientos, Andrés F., Peña Pizarro, Víctor|||0000-0002-3801-5203
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/424845
Acceso en línea:https://hdl.handle.net/2117/424845
https://dx.doi.org/10.1214/19-ba1155
Access Level:acceso abierto
Palabra clave:Bootstrap
Big data
Bayesian nonparametric
Scalable inference
Àrees temàtiques de la UPC::Matemàtiques i estadística
Descripción
Sumario:In this article, we present data-subsetting algorithms that allow for the approximate and scalable implementation of the Bayesian bootstrap. They are analogous to two existing algorithms in the frequentist literature: the bag of little bootstraps (Kleiner et al., 2014) and the subsampled double bootstrap (Sengupta et al., 2016). Our algorithms have appealing theoretical and computational properties that are comparable to those of their frequentist counterparts. Additionally, we provide a strategy for performing lossless inference for a class of functionals of the Bayesian bootstrap and briefly introduce extensions to the Dirichlet Process.