Convergence of Monte Carlo distribution estimates from rival samplers

File Description SizeFormat 
mc_convergence_sac_revision.pdfAccepted version341.38 kBAdobe PDFView/Open
Title: Convergence of Monte Carlo distribution estimates from rival samplers
Authors: Heard, NA
Turcotte, MJM
Item Type: Journal Article
Abstract: It is often necessary to make sampling-based statistical inference about many probability distributions in parallel. Given a finite computational resource, this article addresses how to optimally divide sampling effort between the samplers of the different distributions. Formally approaching this decision problem requires both the specification of an error criterion to assess how well each group of samples represent their underlying distribution, and a loss function to combine the errors into an overall performance score. For the first part, a new Monte Carlo divergence error criterion based on Jensen–Shannon divergence is proposed. Using results from information theory, approximations are derived for estimating this criterion for each target based on a single run, enabling adaptive sample size choices to be made during sampling.
Issue Date: 8-Jul-2015
Date of Acceptance: 1-Jul-2015
ISSN: 1573-1375
Publisher: Springer Verlag
Start Page: 1147
End Page: 1161
Journal / Book Title: Statistics and Computing
Volume: 26
Issue: 6
Copyright Statement: © Springer-Verlag 2015. The final publication is available at Springer via
Keywords: Science & Technology
Physical Sciences
Computer Science, Theory & Methods
Statistics & Probability
Computer Science
Sample sizes
Jensen-Shannon divergence
Transdimensional Markov chains
Statistics & Probability
0104 Statistics
0802 Computation Theory And Mathematics
Publication Status: Published
Appears in Collections:Mathematics
Faculty of Natural Sciences

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx