Discrete chain graph models
From MaRDI portal
Publication:605020
DOI10.3150/08-BEJ172zbMATH Open1452.62348arXiv0909.0843OpenAlexW3106392352WikidataQ57566427 ScholiaQ57566427MaRDI QIDQ605020FDOQ605020
Authors: Mathias Drton 
Publication date: 12 November 2010
Published in: Bernoulli (Search for Journal in Brave)
Abstract: The statistical literature discusses different types of Markov properties for chain graphs that lead to four possible classes of chain graph Markov models. The different models are rather well understood when the observations are continuous and multivariate normal, and it is also known that one model class, referred to as models of LWF (Lauritzen--Wermuth--Frydenberg) or block concentration type, yields discrete models for categorical data that are smooth. This paper considers the structural properties of the discrete models based on the three alternative Markov properties. It is shown by example that two of the alternative Markov properties can lead to non-smooth models. The remaining model class, which can be viewed as a discrete version of multivariate regressions, is proven to comprise only smooth models. The proof employs a simple change of coordinates that also reveals that the model's likelihood function is unimodal if the chain components of the graph are complete sets.
Full work available at URL: https://arxiv.org/abs/0909.0843
Recommendations
algebraic statisticsconditional independencegraphical modelcategorical dataMarkov propertypath diagram
Characterization and structure theory for multivariate probability distributions; copulas (62H05) Random graphs (graph-theoretic aspects) (05C80) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Algebraic statistics (62R01)
Cites Work
- SINGULAR
- Title not available (Why is that?)
- Likelihood ratio tests and singularities
- Graphical models for associations between variables, some of which are qualitative and some quantitative
- Title not available (Why is that?)
- Ancestral graph Markov models.
- Algebraic Statistics for Computational Biology
- Title not available (Why is that?)
- Markov Properties for Acyclic Directed Mixed Graphs
- Marginal models for categorical data
- A SINful approach to Gaussian graphical model selection
- Title not available (Why is that?)
- On the Validity of the Markov Interpretation of Path Diagrams of Gaussian Structural Equations Systems with Correlated Errors
- Stratified exponential families: Graphical models and model selection
- Alternative Markov properties for chain graphs
- Title not available (Why is that?)
- Parameterizations and Fitting of Bi‐directed Graph Models to Categorical Data
- A NOTE ON MULTIVARIATE LOGISTIC MODELS FOR CONTINGENCY TABLES
- Binary Models for Marginal Independence
- Algebraic statistical models
- Regression models for discrete longitudinal responses. With comments and a rejoinder by the authors
- Joint Response Graphs and Separation Induced by Triangular Systems
- Fast estimation algorithm for likelihood-based analysis of repeated categorical responses
- Joint Regression and Association Modeling of Longitudinal Ordinal Data
- Iterative conditional fitting for discrete chain graph models.
- The analysis of multidimensional contingency tables when some variables are posterior to others: a modified path analysis approach
Cited In (37)
- Joint Gaussian graphical model estimation: a survey
- Bayesian learning of multiple directed networks from observational data
- Iterative conditional fitting for discrete chain graph models.
- Model selection and local geometry
- Order-independent structure learning of multivariate regression chain graphs
- AMP chain graphs: minimal separators and structure learning algorithms
- Smooth, identifiable supermodels of discrete DAG models with latent variables
- On the causal interpretation of acyclic mixed graphs under multivariate normality
- Unifying Markov properties for graphical models
- Improving the power of hypothesis tests in sparse contingency tables
- A class of smooth models satisfying marginal and context specific conditional independencies
- Smoothness of marginal log-linear parameterizations
- Discrete models of force chain networks
- On a hypergraph probabilistic graphical model
- Marginalization and conditioning for LWF chain graphs
- Markovian acyclic directed mixed graphs for discrete data
- Marginal parameterizations of discrete models defined by a set of conditional independencies
- Chain graph models of multivariate regression type for categorical data
- Context-specific independencies embedded in chain graph models of type I
- On expressiveness of the chain graph interpretations
- Bayesian Model Selection of Gaussian Directed Acyclic Graph Structures
- Comments on: Sequences of regressions and their independences
- Properties of graphical regression models for multidimensional categorical data
- Sequences of regressions and their independences
- Multiple hidden Markov models for categorical time series
- Every LWF and AMP chain graph originates from a set of causal models
- Triangular systems for symmetric binary variables
- A review of Gaussian Markov models for conditional independence
- Marginal AMP chain graphs
- Graphical models for multivariate Markov chains
- Chain graph interpretations and their relations revisited
- Probability distributions with summary graph structure
- Log-mean linear parameterization for discrete graphical models of marginal independence and the analysis of dichotomizations
- Type II chain graph models for categorical data: a smooth subclass
- Markov properties for mixed graphs
- Context-specific independencies in stratified chain regression graphical models
- A decomposition-based algorithm for learning the structure of multivariate regression chain graphs
Uses Software
This page was built for publication: Discrete chain graph models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q605020)
