Abstract: Log-linear models are a classical tool for the analysis of contingency tables. In particular, the subclass of graphical log-linear models provides a general framework for modelling conditional independences. However, with the exception of special structures, marginal independence hypotheses cannot be accommodated by these traditional models. Focusing on binary variables, we present a model class that provides a framework for modelling marginal independences in contingency tables. The approach taken is graphical and draws on analogies to multivariate Gaussian models for marginal independence. For the graphical model representation we use bi-directed graphs, which are in the tradition of path diagrams. We show how the models can be parameterized in a simple fashion, and how maximum likelihood estimation can be performed using a version of the Iterated Conditional Fitting algorithm. Finally we consider combining these models with symmetry restrictions.
Recommendations
Cites work
- scientific article; zbMATH DE number 992990 (Why is no real title available?)
- scientific article; zbMATH DE number 47310 (Why is no real title available?)
- scientific article; zbMATH DE number 1134987 (Why is no real title available?)
- scientific article; zbMATH DE number 3303654 (Why is no real title available?)
- A different construction of Gaussian fields from Markov chains: Dirichlet covariances.
- A note on the quadratic exponential binary distribution
- Additive and multiplicative models and interactions
- Analogies between Multiplicative Models in Contingency Tables and Covariance Selection
- Ancestral graph Markov models.
- Association Models for a Multivariate Binary Response
- Asymptotically efficient estimation of covariance matrices with linear structure
- Estimation of a covariance matrix with zeros
- Exploring interactions in high-dimensional tables: a bootstrap alternative to log-linear models
- Fast estimation algorithm for likelihood-based analysis of repeated categorical responses
- Introduction to Graphical Modelling
- Joint Regression and Association Modeling of Longitudinal Ordinal Data
- Lancaster interactions revisited
- Linear dependencies represented by chain graphs. With comments and a rejoinder by the authors
- Marginal models for categorical data
- Markov Properties for Acyclic Directed Mixed Graphs
- Markov fields and log-linear interaction models for contingency tables
- Maximum Likelihood Estimation in Gaussian Chain Graph Models under the Alternative Markov Property
- On identification of multi-factor models with correlated residuals
- Polyhedral conditions for the nonexistence of the MLE for hierarchical log-linear models
- Simultaneously Modeling Joint and Marginal Distributions of Multivariate Categorical Responses
- Some equivalence results concerning multiplicative lattice decompositions of multivariate densities
- Stochastic independence, algebraic independence and abstract connectedness
- Stratified exponential families: Graphical models and model selection
- Symmetry and lattice conditional independence in a multivariate normal distribution
- Testing for Pairwise Independence
Cited in
(37)- Bayesian inference in cumulative distribution fields
- Joint Gaussian graphical model estimation: a survey
- Modelling relational statistics with Bayes nets
- Label- and level-invariant graphical log-linear models
- Probability based independence sampler for Bayesian quantitative learning in graphical log-linear marginal models
- Parameterizations and Fitting of Bi‐directed Graph Models to Categorical Data
- Smooth, identifiable supermodels of discrete DAG models with latent variables
- Unifying Markov properties for graphical models
- Marginal log-linear parameterization of conditional independence models
- Graphical modelling of multivariate time series
- Conditional independence in stationary distributions of diffusions
- Modeling conditional and marginal association in contingency tables
- Faithfulness of probability distributions and graphs
- Causal Structural Learning via Local Graphs
- Normal binary graph models
- Hierarchical models for independence structures of networks
- Marginal independence and partial set partitions
- Discrete chain graph models
- Chain graph models of multivariate regression type for categorical data
- Parameter identification in linear non-Gaussian causal models under general confounding
- Confidence intervals for linear unbiased estimators under constrained dependence
- Implicit inequality constraints in a binary tree model
- Graphical and Recursive Models for Contingency Tables
- Modeling sums of exchangeable binary variables
- Harmonic analysis of symmetric random graphs.
- Split models for contingency tables
- Sequences of regressions and their independences
- Marginal log-linear parameters for graphical Markov models
- Conjugate and conditional conjugate Bayesian analysis of discrete graphical models of marginal independence
- Triangular systems for symmetric binary variables
- Exchangeable Bernoulli random variables and Bayes postulate
- Conditional and marginal association for binary random variables
- Log-mean linear models for binary data
- Conjunctive Bayesian networks
- Log-mean linear parameterization for discrete graphical models of marginal independence and the analysis of dichotomizations
- Markov properties for mixed graphs
- A class of log-linear models with constrained marginal distributions
This page was built for publication: Binary Models for Marginal Independence
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3541265)
