Parameterizations and Fitting of Bi‐directed Graph Models to Categorical Data

.  We discuss two parameterizations of models for marginal independencies for discrete distributions which are representable by bi-directed graph models, under the global Markov property. Such models are useful data analytic tools especially if used in combination with other graphical models. The first parameterization, in the saturated case, is also known as thenation multivariate logistic transformation, the second is a variant that allows, in some (but not all) cases, variation-independent parameters. An algorithm for maximum likelihood fitting is proposed, based on an extension of the Aitchison and Silvey method.

[1]  S. Orbom,et al.  When Can Association Graphs Admit A Causal Interpretation? , 1993 .

[2]  J. Aitchison,et al.  Maximum-Likelihood Estimation of Parameters Subject to Restraints , 1958 .

[3]  Tamás Rudas,et al.  On applications of marginal models for categorical data , 2004 .

[4]  G. Kauermann.  On a dualization of graphical Gaussian models , 1996 .

[5]  T. Richardson,et al.  Binary models for marginal independence , 2007, 0707.3794.

[6]  A. Roddam.  An approximate maximum likelihood procedure for parameter estimation in multivariate discrete data regression models , 2001 .

[7]  N. Wermuth,et al.  Covariance chains , 2006 .

[8]  D. Madigan,et al.  Alternative Markov Properties for Chain Graphs , 2001 .

[9]  N. Wermuth,et al.  Linear Dependencies Represented by Chain Graphs , 1993 .

[10]  Tamás Rudas,et al.  Parameterization and estimation of path models for categorical data , 2006 .

[11]  Göran Kauermann,et al.  A NOTE ON MULTIVARIATE LOGISTIC MODELS FOR CONTINGENCY TABLES , 1997 .

[12]  T. Richardson.  Markov Properties for Acyclic Directed Mixed Graphs , 2003 .

[13]  Peter Green,et al.  Highly Structured Stochastic Systems , 2003 .

[14]  G. Molenberghs,et al.  Marginal modelling of multivariate categorical data. , 1999, Statistics in medicine.

[15]  Francesco Bartolucci,et al.  An extended class of marginal link functions for modelling contingency tables by equality and inequality constraints , 2007 .

[16]  A. Coppen.  The Marke-Nyman temperament scale: an English translation. , 1966, The British journal of medical psychology.

[17]  A. Agresti,et al.  Simultaneously Modeling Joint and Marginal Distributions of Multivariate Categorical Responses , 1994 .

[18]  S. Lauritzen,et al.  Globally convergent algorithms for maximizing a likelihood function , 1991 .

[19]  P. Spirtes,et al.  Ancestral graph Markov models , 2002 .

[20]  D. Edwards.  Introduction to graphical modelling , 1995 .

[21]  On the Relation Between with Alternative Codinqs , 1992 .

[22]  M. Drton.  Iterative Conditional Fitting for Discrete Chain Graph Models , 2008 .

[23]  P. McCullagh,et al.  Multivariate Logistic Models , 1995 .

[24]  Joseph B. Lang,et al.  Maximum likelihood methods for a generalized class of log-linear models , 1996 .

[25]  G. M. Marchetti,et al.  Parameterization and Fitting of a Class of Discrete Graphical Models , 2008 .

[26]  A. Forcina,et al.  Marginal regression models for the analysis of positive association of ordinal response variables , 2001 .

[27]  Wicher Bergsma,et al.  Marginal models for dependent, clustered, and longitudinal categorical data , 2009 .

[28]  Nanny Wermuth,et al.  An approximation to maximum likelihood estimates in reduced models , 1990 .

[29]  G. Molenberghs,et al.  Marginal Modeling of Correlated Ordinal Data Using a Multivariate Plackett Distribution , 1994 .

[30]  Judea Pearl,et al.  When can association graphs admit a causal interpretation , 1994 .

[31]  Wicher P. Bergsma,et al.  Marginal models for categorical data , 2002 .

[32]  Dimitri P. Bertsekas,et al.  Constrained Optimization and Lagrange Multiplier Methods , 1982 .

[33]  Michael I. Jordan.  Graphical Models , 2003 .