On the correspondence from Bayesian log-linear modelling to logistic regression modelling with g-priors

DOI10.1007/S11749-017-0540-8zbMATH Open1386.62006arXiv1409.3795OpenAlexW2611901629WikidataQ59614604 ScholiaQ59614604MaRDI QIDQ1708367FDOQ1708367

Michail Papathomas

Publication date: 23 March 2018

Published in: Test (Search for Journal in Brave)

Abstract: Consider a set of categorical variables where at least one of them is binary. The log-linear model that describes the counts in the resulting contingency table implies a specific logistic regression model, with the binary variable as the outcome. Within the Bayesian framework, the $g$-prior and mixtures of $g$-priors are commonly assigned to the parameters of a generalized linear model. We prove that assigning a $g$-prior (or a mixture of $g$-priors) to the parameters of a certain log-linear model designates a $g$-prior (or a mixture of $g$-priors) on the parameters of the corresponding logistic regression. By deriving an asymptotic result, and with numerical illustrations, we demonstrate that when a $g$-prior is adopted, this correspondence extends to the posterior distribution of the model parameters. Thus, it is valid to translate inferences from fitting a log-linear model to inferences within the logistic regression framework, with regard to the presence of main effects and interaction terms.

Full work available at URL: https://arxiv.org/abs/1409.3795

Recommendations

• Prior induction in log-linear models for general contingency table analysis.
• Bayesian inference for Poisson and multinomial log-linear models
• Bayesian selection of log‐linear models
• Bayesian inference for logistic models in R language
• Informative \(g\)-priors for logistic regression

zbMATH Keywords

contingency tablescategorical variablesmixtures of \(g\)-priorsposterior correspondenceprior correspondence

Mathematics Subject Classification ID

Bayesian inference (62F15) Contingency tables (62H17) Generalized linear models (logistic models) (62J12)

Cites Work

• Bayesian Modeling Using WinBUGS
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Mixtures of g Priors for Bayesian Variable Selection
• A fast procedure for model search in multidimensional contingency tables
• Title not available (Why is that?)
• Markov chain Monte Carlo model determination for hierarchical and graphical log-linear models
• Bayesian variable and link determination for generalised linear models
• Title not available (Why is that?)
• A default prior distribution for contingency tables with dependent factor levels
• Title not available (Why is that?)
• A Reference Bayesian Test for Nested Hypotheses and its Relationship to the Schwarz Criterion
• Joint specification of model space and parameter space prior distributions
• Approximate Bayesian model selection with the deviance statistic
• Exploring dependence between categorical variables: benefits and limitations of using variable selection within Bayesian clustering in relation to log-linear modelling with interaction terms
• Hyper-\(g\) priors for generalized linear models
• A novel reversible jump algorithm for generalized linear models
• Power-expected-posterior priors for variable selection in Gaussian linear models
• Graphs and matrices
• Compatibility of prior specifications across linear models
• Title not available (Why is that?)
• Title not available (Why is that?)
• A mixture of \(g\)-priors for variable selection when the number of regressors grows with the sample size

Cited In (5)

• Exploring dependence between categorical variables: benefits and limitations of using variable selection within Bayesian clustering in relation to log-linear modelling with interaction terms
• A weakly informative default prior distribution for logistic and other regression models
• Optimal Gaussian approximations to the posterior for log-linear models with Diaconis-Ylvisaker priors
• A novel Bayesian approach to estimate long memory parameter
• On the use of Cauchy prior distributions for Bayesian logistic regression

Uses Software

• R
• WinBUGS
• gamair
• conting

This page was built for publication: On the correspondence from Bayesian log-linear modelling to logistic regression modelling with \(g\)-priors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1708367)