Generating large Ising models with Markov structure via simple linear relations

We extend the notion of a tree graph to sequences of prime graphs which are cycles and edges and name these non-chordal graphs hollow trees. These structures are especially attractive for palindromic Ising models, which mimic a symmetry of joint Gaussian distributions. We show that for an Ising model all defining independences are captured by zero partial correlations and conditional correlations agree with partial correlations within each prime graph if and only if the model is palindromic and has a hollow-tree structure. This implies that the strength of dependences can be assessed locally. We use the results to find a well-fitting general Ising model with hollow-tree structure for a set of longitudinal data.

[1]  S. Lauritzen,et al.  Maximum likelihood estimation in Gaussian models under total positivity , 2017, The Annals of Statistics.

[2]  Milan Studený,et al.  The Chordal Graph Polytope for Learning Decomposable Models , 2016, Probabilistic Graphical Models.

[3]  N. Wermuth,et al.  Pairwise Markov properties for regression graphs , 2015, 1512.09016.

[4]  Jacques A. Hagenaars,et al.  Methodological Issues in Categorical Data Analysis , 2015 .

[5]  N. Wermuth,et al.  Palindromic Bernoulli distributions , 2015, 1510.09072.

[6]  Shaun M. Fallat,et al.  Total positivity in Markov structures , 2015, 1510.01290.

[7]  Piotr Zwiernik,et al.  Binary distributions of concentric rings , 2014, J. Multivar. Anal..

[8]  N. Wermuth,et al.  Star graphs induce tetrad correlations: for Gaussian as well as for binary variables , 2013, 1307.5396.

[9]  Po-Ling Loh,et al.  Structure estimation for discrete graphical models: Generalized covariance matrices and their inverses , 2012, NIPS.

[10]  N. Wermuth.  Traceable Regressions , 2012 .

[11]  Kayvan Sadeghi,et al.  Graphical Markov Models with Mixed Graphs in R , 2012, R J..

[12]  N. Wermuth,et al.  Sequences of regressions and their independences , 2011, 1110.1986.

[13]  N. Wermuth,et al.  Changing Parameters by Partial Mappings , 2010 .

[14]  M. Drton.  Discrete chain graph models , 2009, 0909.0843.

[15]  N. Wermuth,et al.  Matrix representations and independencies in directed acyclic graphs , 2009, 0904.0333.

[16]  Peter J Green,et al.  Enumerating the Junction Trees of a Decomposable Graph , 2009, Journal of computational and graphical statistics : a joint publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America.

[17]  Vwani P. Roychowdhury,et al.  Covariance selection for nonchordal graphs via chordal embedding , 2008, Optim. Methods Softw..

[18]  Frantisek Matús,et al.  On Gaussian conditional independence structures , 2007, Kybernetika.

[19]  B. Qaqish,et al.  Multivariate logistic models , 2006 .

[20]  Robert Castelo,et al.  A Robust Procedure For Gaussian Graphical Model Search From Microarray Data With p Larger Than n , 2006, J. Mach. Learn. Res..

[21]  N. Wermuth,et al.  Partial inversion for linear systems and partial closure of independence graphs , 2006 .

[22]  Yefim Dinitz,et al.  Dinitz' Algorithm: The Original Version and Even's Version , 2006, Essays in Memory of Shimon Even.

[23]  Søren Højsgaard,et al.  A common platform for graphical models in R , 2005 .

[24]  R. Shibata,et al.  PARTIAL CORRELATION AND CONDITIONAL CORRELATION AS MEASURES OF CONDITIONAL INDEPENDENCE , 2004 .

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

[26]  Nanny Wermuth,et al.  On association models defined over independence graphs , 1998 .

[27]  M. Studený,et al.  On chain graph models for description of conditional independence structures , 1998 .

[28]  G. Glonek.  A class of regression models for multivariate categorical responses , 1996 .

[29]  N. Wermuth,et al.  Tests of Linearity, Multivariate Normality and the Adequacy of Linear Scores , 1994 .

[30]  Nanny Wermuth,et al.  A note on the quadratic exponential binary distribution , 1994 .

[31]  On the Maximum-entropy Extensions of Probability Measures over Undirected Graphs 1 , 1994 .

[32]  Hanns-Georg Leimer,et al.  Optimal decomposition by clique separators , 1993, Discret. Math..

[33]  B. Everitt.  Log-linear models for contingency tables , 1992 .

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

[35]  M. Frydenberg.  The chain graph Markov property , 1990 .

[36]  N. Wermuth,et al.  Graphical Models for Associations between Variables, some of which are Qualitative and some Quantitative , 1989 .

[37]  Juni Palmgren.  Regression Models for Bivariate Binary Responses , 1989 .

[38]  J. E. Jackson.  The Analysis of Cross-Classified Data Having Ordered Categories , 1986 .

[39]  T. Speed,et al.  Gaussian Markov Distributions over Finite Graphs , 1986 .

[40]  P. M. E. Altham,et al.  Improving the Precision of Estimation by Fitting a Model , 1984 .

[41]  Robert E. Tarjan,et al.  Simple Linear-Time Algorithms to Test Chordality of Graphs, Test Acyclicity of Hypergraphs, and Selectively Reduce Acyclic Hypergraphs , 1984, SIAM J. Comput..

[42]  S. Karlin,et al.  M-Matrices as covariance matrices of multinormal distributions☆ , 1983 .

[43]  S. M. Hedetniemi,et al.  Linear Algorithms for Finding the Jordan Center and Path Center of a Tree , 1981 .

[44]  A. Dawid.  Conditional Independence in Statistical Theory , 1979 .

[45]  Stephen E. Fienberg,et al.  Discrete Multivariate Analysis: Theory and Practice , 1976 .

[46]  J. Besag.  Spatial Interaction and the Statistical Analysis of Lattice Systems , 1974 .

[47]  D. Rose.  Triangulated graphs and the elimination process , 1970 .

[48]  A. W. Edwards.  The Measure of Association in a 2 × 2 Table , 1963 .

[49]  M. W. Birch.  Maximum Likelihood in Three-Way Contingency Tables , 1963 .

[50]  Irving John Good,et al.  The Interaction Algorithm and Practical Fourier Analysis , 1958 .

[51]  R. Fisher,et al.  On the Mathematical Foundations of Theoretical Statistics , 1922 .