|
|||||
|
|
Componentwise adaptation for high dimensional MCMC |
|
| Authors: | Heikki Haario Eero Saksman Johanna Tamminen |
| Affiliation: | 1. University of Helsinki Department of Mathematics and Statistics, University of Helsinki, P.O.Box 4, FIN-00014, Finland 2. University of Jyv?skyl? Department of Mathematics and Statistics, University of Jyv?skyl?, P.O.Box 35 (MaD), FIN-40014, Finland 3. Finnish Meteorological Institute Geophysical Research Division, P.O.Box 503, FIN-00101, Helsinki, Finland |
| Abstract: | Summary We introduce a new adaptive MCMC algorithm, based on the traditional single component Metropolis-Hastings algorithm and on our earlier adaptive Metropolis algorithm (AM). In the new algorithm the adaption is performed component by component. The chain is no more Markovian, but it remains ergodic. The algorithm is demonstrated to work well in varying test cases up to 1000 dimensions. |
| Keywords: | MCMC adaptive MCMC Metropolis-Hastings algorithm |
| 本文献已被 SpringerLink 等数据库收录！ | |