Parallel inference for big data with the group Bayesian method

DOI10.1007/S00184-020-00784-0MaRDI QIDQ2227203zbMATH OpenOpenAlexFDO

Authors Guangbao Guo, Guoqi Qian, Lu Lin, Wei Shao

Publication date 10 February 2021

Published in Metrika (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s00184-020-00784-0

zbMATH Keywords

data subsetsgroup Gibbsparallel inference

Mathematics Subject Classification ID

Sampling theory, sample surveys (62D05) Bayesian inference (62F15) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Generalized linear models (logistic models) (62J12) Statistical aspects of big data and data science (62R07)

Review text

The group Gibbs method is employed to derive the regression estimator. The estimator variance is reduced using a weighted estimator and posterior. The convergence is studied from the theoretical perspective. The existence of an optimal weight sequence is established. Numerical simulations have demonstrated that the proposed approach is effective comparing with conventional Markov chain Monte Carlo methods.

Recommendations

• Parallel Markov chain Monte Carlo for Bayesian hierarchical models with big data, in two stages
• Double-parallel Monte Carlo for Bayesian analysis of big data
• A partitioned quasi-likelihood for distributed statistical inference
• Big data Bayesian linear regression and variable selection by normal-inverse-gamma summation
• Parallel statistical computing for statistical inference

Cites work

• A split-and-merge Bayesian variable selection approach for ultrahigh dimensional regression
• Bayesian Subset Modeling for High-Dimensional Generalized Linear Models
• Bayesian variable selection for high dimensional generalized linear models: convergence rates of the fitted densities
• Communication-efficient distributed statistical inference
• Communication-efficient sparse regression
• Parallel MCMC with generalized elliptical slice sampling
• Scalable inference for Markov processes with intractable likelihoods
• Sequential Monte Carlo on large binary sampling spaces
• Speeding Up MCMC by Efficient Data Subsampling
• Statistical methods and computing for big data

Cited in

(9)

• Robust and parallel Bayesian model selection
• Statistical inference in massive datasets by empirical likelihood
• A partitioned quasi-likelihood for distributed statistical inference
• Double-parallel Monte Carlo for Bayesian analysis of big data
• Parallel Markov chain Monte Carlo for Bayesian hierarchical models with big data, in two stages
• A selective review of statistical methods using calibration information from similar studies
• Individualized Group Learning
• The COR criterion for optimal subset selection in distributed estimation
• Cor: an R package for optimal subset selection in distributed estimation

Describes a project that uses

Uses Software

• runjags

This page was built for publication: Parallel inference for big data with the group Bayesian method

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2227203)