Improving the efficiency and robustness of nested sampling using posterior repartitioning
- Submitting institution
-
The University of Bath
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 203142382
- Type
- D - Journal article
- DOI
-
10.1007/s11222-018-9841-3
- Title of journal
- Statistics and Computing
- Article number
- -
- First page
- 835
- Volume
- 29
- Issue
- 4
- ISSN
- 0960-3174
- Open access status
- Compliant
- Month of publication
- November
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The unrepresentative prior problem occurs when prior assumptions regarding parameters of interest are unrepresentative of their actual values for a given dataset. This may lead to inefficient exploration of the resulting posterior in a sampling algorithm. This paper introduces a posterior repartitioning method which significantly increases the efficiency of sampling while keeping posterior inferences and evidence estimates unchanged. This work was motivated by practical industrial needs and was developed as part of the machine learning solution in collaboration with Shell, a top company in the oil/gas industry.
Contact: Principal Reservoir Engineer at Shell
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -