Bayesian Inference from Composite Likelihoods, with an Application to\n Spatial Extremes

2009/11/27 by Mathieu Ribatet, Daniel Cooley, Ribatet, Mathieu +3 · 8 citations
Environmental Science · Mathematics · #FOS: Computer and information sciences #FOS: Mathematics #Hydrology and Drought Analysis #Methodology (stat.ME) #Soil Geostatistics and Mapping #Statistical Methods and Bayesian Inference #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.0911.5357

openalex publication_date 2009/11/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Composite likelihoods are increasingly used in applications where the full\nlikelihood is analytically unknown or computationally prohibitive. Although the\nmaximum composite likelihood estimator has frequentist properties akin to those\nof the usual maximum likelihood estimator, Bayesian inference based on\ncomposite likelihoods has yet to be explored. In this paper we investigate the\nuse of the Metropolis--Hastings algorithm to compute a pseudo-posterior\ndistribution based on the composite likelihood. Two methodologies for adjusting\nthe algorithm are presented and their performance on approximating the true\nposterior distribution is investigated using simulated data sets and real data\non spatial extremes of rainfall.\n

Cited by

• Copulas for Geostatistical Data: Foundations, Modeling Principles and Statistical Inference
• Differentially Private Bayesian Inference for Gaussian Copula Correlations
• Fast and Accurate Node-Age Estimation Under Fossil Calibration Uncertainty Using the Adjusted Pairwise Likelihood
• Uncertainty Quantification for Multi-level Models Using the Survey-Weighted Pseudo-Posterior
• Bayes Factor Tests for Group Differences in Ordinal and Binary Graphical Models
• Uncertainty Estimation for Pseudo‐Bayesian Inference Under Complex Sampling
• Efficient Bayesian Inference for Spatial Point Patterns Using the Palm Likelihood
• Calibration of conditional composite likelihood for Bayesian inference on Gibbs random fields

Related

• Approximate Bayesian Computing for Spatial Extremes
• Reliable inference for complex models by discriminative composite likelihood estimation
• High-order Composite Likelihood Inference for Max-Stable Distributions\n and Processes
• Optimal combination of composite likelihoods using approximate Bayesian computation with application to state-space models
• Bayesian inference for Gibbs random fields using composite likelihoods