glabcmcmc : a Python package for ABC-MCMC with local and global moves

We introduce a new Python package glabcmcmc, which implements an approximate Bayesian computation Markov chain Monte Carlo (ABC-MCMC) algorithm that combines global and local proposal strategies to address the limitations of standard ABC-MCMC. The proposed package includes key innovations such as th...

Full description

Saved in:
Bibliographic Details
Published inStatistical theory and related fields Vol. 9; no. 2; pp. 168 - 177
Main Authors Cao, Xuefei, Wang, Shijia, Zhou, Yongdao
Format Journal Article
LanguageEnglish
Published Taylor & Francis Group 03.04.2025
Subjects
Approximate Bayesian Computation
global-local proposal
Markov chain Monte Carlo
Online AccessGet full text

Cover

More Information
Summary:We introduce a new Python package glabcmcmc, which implements an approximate Bayesian computation Markov chain Monte Carlo (ABC-MCMC) algorithm that combines global and local proposal strategies to address the limitations of standard ABC-MCMC. The proposed package includes key innovations such as the determination of global proposal frequencies, the implementation of a hybrid ABC-MCMC algorithm integrating global and local proposals, and an adaptive version that utilizes normalizing flows and gradient-based computations for enhanced proposal mechanisms. The functionality of the software package is demonstrated through illustrative examples.
ISSN:2475-4269
2475-4277
DOI:10.1080/24754269.2025.2495505