Langevin Algorithms : Questions at the Numerical Analysis / Applied Probability Interface

Lead Research Organisation: University of Warwick
Department Name: Statistics

Abstract

The aim of the proposal is to develop and analyze a broad and effective class of MCMC methods, termed Langevin algorithms; in particular we aim to cross-fertilize between the fields of Numerical Analysis and Applied Probability to obtain potentially significant benefits in both the understanding of existing algorithms, and in the development of new ones.

Publications

10 25 50
publication icon
Papaspiliopoulos O (2013) Data Augmentation for Diffusions in Journal of Computational and Graphical Statistics

publication icon
Peluchetti S (2013) The strong weak convergence of the quasi-EA in Queueing Systems

Related Projects

Project Reference Relationship Related To Start End Award Value
EP/D505607/1 01/02/2006 30/04/2007 £124,991
EP/D505607/2 Transfer EP/D505607/1 01/05/2007 31/07/2008 £69,679