EUVIMEDEUVIMED entdecken
Crossref

Revisiting the Gelman–Rubin Diagnostic

Dootika Vats, Christina Knudson

Statistical Science · 01.11.2021 · 36 (4) · 518-529

DOI: 10.1214/20-sts812 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Statistical Science, erschienen am 01.11.2021. Verfasst von Dootika Vats und 1 weiteren Autor:innen. Die Arbeit wurde bisher 144-mal zitiert (Crossref) und nennt 37 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Gelman and Rubin’s (Statist. Sci. 7 (1992) 457–472) convergence diagnostic is one of the most popular methods for terminating a Markov chain Monte Carlo (MCMC) sampler. Since the seminal paper, researchers have developed sophisticated methods for estimating variance of Monte Carlo averages. We show that these estimators find immediate use in the Gelman–Rubin statistic, a connection not previously established in the literature. We incorporate these estimators to upgrade both the univariate and multivariate Gelman–Rubin statistics, leading to improved stability in MCMC termination time. An immediate advantage is that our new Gelman–Rubin statistic can be calculated for a single chain. In addition, we establish a one-to-one relationship between the Gelman–Rubin statistic and effective sample size. Leveraging this relationship, we develop a principled termination criterion for the Gelman–Rubin statistic. Finally, we demonstrate the utility of our improved diagnostic via examples.

Zitieren

APA

Vats, D., & Knudson, C. (2021). Revisiting the Gelman–Rubin Diagnostic. Statistical Science, 36(4), 518-529. https://doi.org/10.1214/20-sts812

Vancouver

Vats D, Knudson C. Revisiting the Gelman–Rubin Diagnostic. Statistical Science. 2021;36(4):518-529. doi:10.1214/20-sts812

BibTeX
@article{Vats2021,
  author = {Vats, Dootika and Knudson, Christina},
  title = {Revisiting the Gelman–Rubin Diagnostic},
  journal = {Statistical Science},
  year = {2021},
  volume = {36},
  number = {4},
  pages = {518--529},
  doi = {10.1214/20-sts812}
}
Kostenlos starten →