Semi-Exact Control Functionals From Sard's Method

South, Leah and Karvonen, Toni and Nemeth, Christopher and Girolami, Mark and Oates, Chris J. (2022) Semi-Exact Control Functionals From Sard's Method. Biometrika, 109 (2). pp. 351-367. ISSN 0006-3444

Text (Bayes_Sard_Stein)
Bayes_Sard_Stein.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial.
Download (479kB)

Abstract

A novel control variate technique is proposed for post-processing of Markov chain Monte Carlo output, based both on Stein's method and an approach to numerical integration due to Sard. The resulting estimators of posterior expected quantities of interest are proven to be polynomially exact in the Gaussian context, while empirical results suggest the estimators approximate a Gaussian cubature method near the Bernstein-von-Mises limit. The main theoretical result establishes a bias-correction property in settings where the Markov chain does not leave the posterior invariant. Empirical results are presented across a selection of Bayesian inference tasks. All methods used in this paper are available in the R package ZVCV.

Item Type:

Journal Article

Journal or Publication Title:

Biometrika

Additional Information:

This is a pre-copy-editing, author-produced PDF of an article accepted for publication in Biometrika following peer review. The definitive publisher-authenticated version L F South, T Karvonen, C Nemeth, M Girolami, C J Oates, Semi-exact control functionals from Sard’s method, Biometrika, Volume 109, Issue 2, June 2022, Pages 351–367 is available online at: https://academic.oup.com/biomet/article/109/2/351/6309456

Uncontrolled Keywords:

/dk/atira/pure/subjectarea/asjc/1100

Subjects:

?? agricultural and biological sciences(all)applied mathematicsstatistics and probabilitystatistics, probability and uncertaintymathematics(all)agricultural and biological sciences (miscellaneous) ??

Departments:

Faculty of Science and Technology > Mathematics and Statistics

ID Code:

155711

Deposited By:

ep_importer_pure

Deposited On:

03 Jun 2021 11:10

Refereed?:

Yes

Published?:

Published

Last Modified:

17 Dec 2023 01:52

URI:

https://eprints.lancs.ac.uk/id/eprint/155711