Detecting and diagnosing prior and likelihood sensitivity with power-scaling

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.

31 pages, 15 (+5 suppl) figures

References in corpus (3)

• The prior can generally only be understood in the context of the likelihood
• Graphical Test for Discrete Uniformity and its Applications in Goodness of Fit Evaluation and Multiple Sample Comparison
• Robustness against conflicting prior information in regression

Cited by in corpus (7)

• Uncertainty Quantification and Propagation in Surrogate-based Bayesian Inference
• Identifying supermassive black hole recoil in elliptical galaxies
• Hierarchical Bayesian Models to Mitigate Systematic Disparities in Prediction with Proxy Outcomes
• Bayesian random-effects meta-analysis of aggregate data on clinical events
• Stellar Cores Live Long and Prosper in Cuspy Dark Matter Halos
• The ARR2 prior: flexible predictive prior definition for Bayesian auto-regressions
• Power-scaled Bayesian Inference with Score-based Generative Models