Generalised likelihood profiles for models with intractable likelihoods

DOI10.1007/S11222-023-10361-WMaRDI QIDQ6190673zbMATH OpenFDO

Authors David J. Warne, Oliver J. Maclaren, Elliot J. Carr, Matthew J. Simpson, C. C. Drovandi

Publication date 6 February 2024

Published in Statistics and Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2305.10710

zbMATH Keywords

parametric bootstrapintractable likelihoodsimulation-based inferenceprofilinggeneralised likelihood

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Point estimation (62F10) Bootstrap, jackknife and other resampling methods (62F40)

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.

Recommendations

• A simple approach to maximum intractable likelihood estimation
• Statistics for model calibration
• Bayesian parametric bootstrap for models with intractable likelihoods
• Parameter estimation for hidden Markov models with intractable likelihoods
• Integration based profile likelihood calculation for PDE constrained parameter estimation problems

Cites work

• scientific article; zbMATH DE number 3163319 (Why is no real title available?)
• scientific article; zbMATH DE number 3925957 (Why is no real title available?)
• scientific article; zbMATH DE number 46578 (Why is no real title available?)
• scientific article; zbMATH DE number 47310 (Why is no real title available?)
• scientific article; zbMATH DE number 3458075 (Why is no real title available?)
• scientific article; zbMATH DE number 1122119 (Why is no real title available?)
• scientific article; zbMATH DE number 1943901 (Why is no real title available?)
• scientific article; zbMATH DE number 1782877 (Why is no real title available?)
• A General Framework for Updating Belief Distributions
• A Scalable Bootstrap for Massive Data
• A flexible regression model for count data
• A local-global multiscale method for highly heterogeneous stochastic groundwater flow problems
• A trust region method based on interior point techniques for nonlinear programming.
• An Optimum Property of Regular Maximum Likelihood Estimation
• Bayesian Synthetic Likelihood
• Bootstrap methods: another look at the jackknife
• Calibrating general posterior credible regions
• Empirical likelihood ratio confidence intervals for a single functional
• Empirical likelihood ratio confidence regions
• Handbook of approximate Bayesian computation
• Hierarchical Bayesian spatio-temporal Conway-Maxwell Poisson models with dynamic dispersion
• Inference based on estimating functions in the presence of nuisance parameters. With comments and rejoinder
• Inference functions and quadratic score tests
• Invariance of Maximum Likelihood Estimators
• Likelihood methods in statistics
• Likelihood ratio tests and singularities
• Maximum Likelihood: An Introduction
• Minimum Distance and Robust Estimation
• Modified quasi-profile likelihoods from estimating functions
• On the asymptotics of constrained local M-estimators.
• Particle Markov Chain Monte Carlo Methods
• Potential functions and conservative estimating functions
• Principles of Statistical Inference
• Rapid Bayesian Inference for Expensive Stochastic Models
• Reliable and efficient parameter estimation using approximate continuum limit descriptions of stochastic models
• Robust Estimation of a Location Parameter
• Robust Statistics
• Science and Statistics
• Statistical Evidence
• The Minimum Distance Method
• The frontier of simulation-based inference
• The theory and applications of statistical inference functions
• Uncertainty quantification in scientific machine learning: methods, metrics, and comparisons

Cited in

(3)

• Making predictions using poorly identified mathematical models
• Likelihood-free frequentist inference: bridging classical statistics and machine learning for reliable simulator-based inference
• Mathematical modelling and uncertainty quantification for analysis of biphasic coral reef recovery patterns

This page was built for publication: Generalised likelihood profiles for models with intractable likelihoods

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6190673)