AUEB Stats Seminars 10/12/2021: Variance reduction for MCMC estimators by Angelos Alexopoulos (University of Cambridge)

Το επόμενο σεμινάριο του Τμήματος Στατιστικής για την Παρασκευή 10/12/2021

Presenter: Angelos Alexopoulos
Research Associate, MRC Biostatistics Unit, University of Cambridge, UK

Date: Friday 10/12/2021, 13:30

Title: Variance reduction for MCMC estimators

Abstract
We provide a general methodology to construct control variates for any discrete time random walk Metropolis and Metropolis-adjusted Langevin algorithm. By adopting the  proposed methods we achieve, in a post-processing manner and with a negligible additional computational cost, impressive variance reduction when compared to the standard MCMC ergodic averages. Our proposed estimators are based on an approximate solution of the Poisson equation for multivariate Gaussian target densities of any dimension.

This is joint work with P. Dellaportas (UCL and AUEB) and M. Titsias (DeepMind)

Link: https://teams.microsoft.com/l/meetup-join/19%3a0fe3bd7e094a4ccfacecb92ce36cfe69%40thread.tacv2/1631393377331?context=%7b%22Tid%22%3a%22ad5ba4a2-7857-4ea1-895e-b3d5207a174f%22%2c%22Oid%22%3a%22381c5dc1-ed77-403a-9d53-e2db51c33563%22%7d