[Forum SIS] seminar: "Revisiting the Gelman-Rubin Diagnostic", Dootika Vats November 9
Umberto Picchini
umberto.picchini a gmail.com
Ven 5 Nov 2021 10:57:40 CET
You are welcome to the next Statistics seminar at Dept of Mathematical
Sciences at Chalmers and Göteborg University.
We are very glad to have, on Tuesday 9 November,Dootika Vats (Indian
Institute of Technology, Kanpur) who will talk of:
/"Revisiting the Gelman-Rubin Diagnostic"/
//******
***
***
***
About the speaker: Dootika Vats <https://dvats.github.io/> is Assistan
Professor at the Indian Instute of Technology in Kanpur. She is the 2021
recipient of the ISBA "Blackwell-Rosenbluth Award". She works in the
general area of Bayesian computation and specifically focus on Markov
chain Monte Carlo algorithms.
****Dootika VatsAbstract *Gelman and Rubin's (1992) 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 of variance estimation for Monte
Carlo averages. We show that this class of estimators find immediate use
in the Gelman-Rubin statistic, a connection not established in the
literature before. We incorporate these estimators to upgrade both the
univariate and multivariate Gelman-Rubin statistics, leading to
increased stability in MCMC termination time. An immediate advantage is
that our new Gelman-Rubin statistic can also be calculated for a single
chain. In addition, we establish a relationship between the Gelman-Rubin
statistic and effective sample size. Leveraging this relationship, we
develop a principled cutoff criterion for the Gelman-Rubin statistic.
Finally, we demonstrate the utility of our improved diagnostic via an
example. This work is joint with Christina Knudson, University of St.
Thomas, Minnesota.** *
*
*
*Feel free to spread this announcement in your network.*
*
*
*Where*: *room MVL15 and zoom* (physical attendance is very welcome)
https://chalmers.zoom.us/j/67043486361
Password: 946764
**
****
*When*: Tuesday 9 November at 14.15-15.15 (Swedish time).
****
****
**
*
--
_________________________________________________________
Umberto Picchini, Associate Professor, PhD, Docent
https://umbertopicchini.github.io/ , twitter: @uPicchini
-------------- parte successiva --------------
Un allegato HTML č stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20211105/4d0abee2/attachment.html>
-------------- parte successiva --------------
Un allegato non testuale č stato rimosso....
Nome: DV_office_crop.jpg
Tipo: image/jpeg
Dimensione: 167430 bytes
Descrizione: non disponibile
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20211105/4d0abee2/attachment.jpg>
Maggiori informazioni sulla lista
Sis