We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.CO

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Computation

Title: Double Happiness: Enhancing the Coupled Gains of L-lag Coupling via Control Variates

Abstract: The recently proposed L-lag coupling for unbiased Markov chain Monte Carlo (MCMC) calls for a joint celebration by MCMC practitioners and theoreticians. For practitioners, it circumvents the thorny issue of deciding the burn-in period or when to terminate an MCMC sampling process, and opens the door for safe parallel implementation. For theoreticians, it provides a powerful tool to establish elegant and easily estimable bounds on the exact error of an MCMC approximation at any finite number of iterates. A serendipitous observation about the bias-correcting term leads us to introduce naturally available control variates into the L-lag coupling estimators. In turn, this extension enhances the coupled gains of L-lag coupling, because it results in more efficient unbiased estimators, as well as a better bound on the total variation error of MCMC iterations, albeit the gains diminish as L increases. Specifically, the new upper bound is theoretically guaranteed to never exceed the one given previously. We also argue that L-lag coupling represents a coupling for the future, breaking from the coupling-from-the-past type of perfect sampling, by reducing the generally unachievable requirement of being perfect to one of being unbiased, a worthwhile trade-off for ease of implementation in most practical situations. The theoretical analysis is supported by numerical experiments that show tighter bounds and a gain in efficiency when control variates are introduced.
Subjects: Computation (stat.CO)
Cite as: arXiv:2008.12662 [stat.CO]
  (or arXiv:2008.12662v4 [stat.CO] for this version)

Submission history

From: Radu V. Craiu [view email]
[v1] Fri, 28 Aug 2020 14:00:47 GMT (355kb,D)
[v2] Mon, 12 Oct 2020 22:55:11 GMT (289kb,D)
[v3] Tue, 2 Feb 2021 16:04:38 GMT (84kb,D)
[v4] Wed, 14 Apr 2021 00:21:04 GMT (81kb,D)

Link back to: arXiv, form interface, contact.