Beskos, A;
Roberts, G;
(2005)
One-shot CFTP; Application to a class of truncated Gaussian densities.
Methodology and Computing in Applied Probability
, 7
pp. 407-437.
10.1007/s11009-005-5001-2.
Preview |
Text
Beskos_BeskosRoberts05.pdf - Accepted Version Download (320kB) | Preview |
Abstract
In this paper, we introduce a new method for perfect simulation of multivariate densities. We use One-Shot CFTP (G. Roberts and J. Rosenthal, “One-shot coupling for certain stochastic recursive sequences,” Stochastic Processes and their Applications vol. 99 pp. 195–208, 2002) together with a monotone coupler for the Gibbs sampler, and implement the algorithm within the Read-Once CFTP protocol (D. B. Wilson, “How to couple form the past using a read-once source of randomness,” Random Structures and Algorithms vol. 16(1) pp. 85–113, 2000b). We illustrate our method by simulating efficiently from high-dimensional truncated normal distributions using the Gibbs sampler.
Type: | Article |
---|---|
Title: | One-shot CFTP; Application to a class of truncated Gaussian densities |
Open access status: | An open access version is available from UCL Discovery |
DOI: | 10.1007/s11009-005-5001-2 |
Publisher version: | https://doi.org/10.1007/s11009-005-5001-2 |
Language: | English |
Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
Keywords: | one-shot coupling, read-once coupling from the past,truncated gaussian densities |
UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Statistical Science |
URI: | https://discovery.ucl.ac.uk/id/eprint/10050410 |
Archive Staff Only
View Item |