Robust adaptive Metropolis algorithm with coerced acceptance rate
The adaptive Metropolis (AM) algorithm of Haario, Saksman and Tamminen (Bernoulli 7(2):223-242, 2001) uses the estimated covariance of the target distribution in the proposal distribution. This paper introduces a new robust adaptive Metropolis algorithm estimating the shape of the target distributio...
| Main Author: | |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
2012
|
| _version_ | 1826306359826579456 |
|---|---|
| author | Vihola, M |
| author_facet | Vihola, M |
| author_sort | Vihola, M |
| collection | OXFORD |
| description | The adaptive Metropolis (AM) algorithm of Haario, Saksman and Tamminen (Bernoulli 7(2):223-242, 2001) uses the estimated covariance of the target distribution in the proposal distribution. This paper introduces a new robust adaptive Metropolis algorithm estimating the shape of the target distribution and simultaneously coercing the acceptance rate. The adaptation rule is computationally simple adding no extra cost compared with the AM algorithm. The adaptation strategy can be seen as a multidimensional extension of the previously proposed method adapting the scale of the proposal distribution in order to attain a given acceptance rate. The empirical results show promising behaviour of the new algorithm in an example with Student target distribution having no finite second moment, where the AM covariance estimate is unstable. In the examples with finite second moments, the performance of the new approach seems to be competitive with the AM algorithm combined with scale adaptation. © 2011 Springer Science+Business Media, LLC. |
| first_indexed | 2024-03-07T06:46:46Z |
| format | Journal article |
| id | oxford-uuid:fb1c345c-54fa-4856-8c3b-f1aaa3079d0d |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T06:46:46Z |
| publishDate | 2012 |
| record_format | dspace |
| spelling | oxford-uuid:fb1c345c-54fa-4856-8c3b-f1aaa3079d0d2022-03-27T13:11:23ZRobust adaptive Metropolis algorithm with coerced acceptance rateJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:fb1c345c-54fa-4856-8c3b-f1aaa3079d0dEnglishSymplectic Elements at Oxford2012Vihola, MThe adaptive Metropolis (AM) algorithm of Haario, Saksman and Tamminen (Bernoulli 7(2):223-242, 2001) uses the estimated covariance of the target distribution in the proposal distribution. This paper introduces a new robust adaptive Metropolis algorithm estimating the shape of the target distribution and simultaneously coercing the acceptance rate. The adaptation rule is computationally simple adding no extra cost compared with the AM algorithm. The adaptation strategy can be seen as a multidimensional extension of the previously proposed method adapting the scale of the proposal distribution in order to attain a given acceptance rate. The empirical results show promising behaviour of the new algorithm in an example with Student target distribution having no finite second moment, where the AM covariance estimate is unstable. In the examples with finite second moments, the performance of the new approach seems to be competitive with the AM algorithm combined with scale adaptation. © 2011 Springer Science+Business Media, LLC. |
| spellingShingle | Vihola, M Robust adaptive Metropolis algorithm with coerced acceptance rate |
| title | Robust adaptive Metropolis algorithm with coerced acceptance rate |
| title_full | Robust adaptive Metropolis algorithm with coerced acceptance rate |
| title_fullStr | Robust adaptive Metropolis algorithm with coerced acceptance rate |
| title_full_unstemmed | Robust adaptive Metropolis algorithm with coerced acceptance rate |
| title_short | Robust adaptive Metropolis algorithm with coerced acceptance rate |
| title_sort | robust adaptive metropolis algorithm with coerced acceptance rate |
| work_keys_str_mv | AT viholam robustadaptivemetropolisalgorithmwithcoercedacceptancerate |