Generalised likelihood profiles for models with intractable likelihoods
arXiv:2305.10710 · doi:10.1007/s11222-023-10361-w
Abstract
Likelihood profiling is an efficient and powerful frequentist approach for parameter estimation, uncertainty quantification and practical identifiablity analysis. Unfortunately, these methods cannot be easily applied for stochastic models without a tractable likelihood function. Such models are typical in many fields of science, rendering these classical approaches impractical in these settings. To address this limitation, we develop a new approach to generalising the methods of likelihood profiling for situations when the likelihood cannot be evaluated but stochastic simulations of the assumed data generating process are possible. Our approach is based upon recasting developments from generalised Bayesian inference into a frequentist setting. We derive a method for constructing generalised likelihood profiles and calibrating these profiles to achieve desired frequentist coverage for a given coverage level. We demonstrate the performance of our method on realistic examples from the literature and highlight the capability of our approach for the purpose of practical identifability analysis for models with intractable likelihoods.
References in corpus (13)
- Uncertainty Quantification in Scientific Machine Learning: Methods, Metrics, and Comparisons
- A flexible regression model for count data
- A General Framework for Updating Belief Distributions
- Likelihood ratio tests and singularities
- Calibrating general posterior credible regions
- Simulation-Based Inference of Reionization Parameters From 3D Tomographic 21 cm Lightcone Images
- Robust Simulation-Based Inference in Cosmology with Bayesian Neural Networks
- A practical guide to pseudo-marginal methods for computational inference in systems biology
- Rapid Bayesian inference for expensive stochastic models
- Importance sampling for stochastic quantum simulations
- Profile likelihood analysis for a stochastic model of diffusion in heterogeneous media
- New homogenization approaches for stochastic transport through heterogeneous media
- A local-global generalized multiscale finite element method for highly heterogeneous stochastic groundwater flow problems