Back to Search
Start Over
Detecting and diagnosing prior and likelihood sensitivity with power-scaling.
- Source :
- Statistics & Computing; Feb2024, Vol. 34 Issue 1, p1-27, 27p
- Publication Year :
- 2024
-
Abstract
- Determining the sensitivity of the posterior to perturbations of the prior and likelihood is an important part of the Bayesian workflow. We introduce a practical and computationally efficient sensitivity analysis approach using importance sampling to estimate properties of posteriors resulting from power-scaling the prior or likelihood. On this basis, we suggest a diagnostic that can indicate the presence of prior-data conflict or likelihood noninformativity and discuss limitations to this power-scaling approach. The approach can be easily included in Bayesian workflows with minimal effort by the model builder and we present an implementation in our new R package priorsense. We further demonstrate the workflow on case studies of real data using models varying in complexity from simple linear models to Gaussian process models. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 09603174
- Volume :
- 34
- Issue :
- 1
- Database :
- Complementary Index
- Journal :
- Statistics & Computing
- Publication Type :
- Academic Journal
- Accession number :
- 174551811
- Full Text :
- https://doi.org/10.1007/s11222-023-10366-5