Hamiltonian Monte Carlo for hierarchical models
File(s)1312.0906v1.pdf (1.68 MB)
Accepted version
Author(s)
Betancourt, MJ
Girolami, Mark
Type
Chapter
Abstract
Hierarchical modeling provides a framework for modeling the complex
interactions typical of problems in applied statistics. By capturing these
relationships, however, hierarchical models also introduce distinctive
pathologies that quickly limit the efficiency of most common methods of in-
ference. In this paper we explore the use of Hamiltonian Monte Carlo for
hierarchical models and demonstrate how the algorithm can overcome those
pathologies in practical applications.
interactions typical of problems in applied statistics. By capturing these
relationships, however, hierarchical models also introduce distinctive
pathologies that quickly limit the efficiency of most common methods of in-
ference. In this paper we explore the use of Hamiltonian Monte Carlo for
hierarchical models and demonstrate how the algorithm can overcome those
pathologies in practical applications.
Editor(s)
Upandhyay, Satyanshu Kumar
Singh, Umesh
Dey, Dipak K
Loganathan, Appaia
Date Issued
2015
Citation
Current Trends in Bayesian Methodology with Applications, 2015, pp.79-100
ISBN
9781482235128
Publisher
Taylor & Francis Group
Start Page
79
End Page
100
Journal / Book Title
Current Trends in Bayesian Methodology with Applications
Copyright Statement
© 2015 by Taylor & Francis Group, LLC. This is an Accepted Manuscript of a book chapter published by Routledge/CRC Press in Current Trends in Bayesian Methodology with Applications in 2015.
Identifier
http://arxiv.org/abs/1312.0906v1
Subjects
stat.ME
stat.ME
Notes
11 pages, 12 figures
Publication Status
Published