Multi-Trek Separation in Linear Structural Equation Models

Building on the theory of causal discovery from observational data, we study interactions between multiple (sets of) random variables in a linear structural equation model with non-Gaussian error terms. We give a correspondence between structure in the higher order cumulants and combinatorial structure in the causal graph. It has previously been shown that low rank of the covariance matrix corresponds to trek separation in the graph. Generalizing this criterion to multiple sets of vertices, we characterize when determinants of subtensors of the higher order cumulant tensors vanish. This criterion applies when hidden variables are present as well. For instance, it allows us to identify the presence of a hidden common cause of k of the observed variables.

[1]  Seth Sullivant,et al.  Positivity for Gaussian graphical models , 2012, Adv. Appl. Math..

[2]  S. Wright.  The Method of Path Coefficients , 1934 .

[3]  M. Drton,et al.  High-dimensional causal discovery under non-Gaussianity , 2018, Biometrika.

[4]  D. A. Kenny,et al.  Correlation and Causation , 1937, Wilmott.

[5]  Judea Pearl,et al.  Fusion, Propagation, and Structuring in Belief Networks , 1986, Artif. Intell..

[6]  S. Gill Williamson,et al.  A Comprehensive Introduction to Linear Algebra , 1989 .

[7]  Peter Bühlmann,et al.  Estimating High-Dimensional Directed Acyclic Graphs with the PC-Algorithm , 2007, J. Mach. Learn. Res..

[8]  Aapo Hyvärinen,et al.  DirectLiNGAM: A Direct Method for Learning a Linear Non-Gaussian Structural Equation Model , 2011, J. Mach. Learn. Res..

[9]  Seth Sullivant,et al.  Algebraic statistics , 2018, ISSAC.

[10]  S. Sullivant,et al.  Trek separation for Gaussian graphical models , 2008, 0812.1938.

[11]  K. Menger.  Zur allgemeinen Kurventheorie , 1927 .

[12]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[13]  Elina Robeva,et al.  Nested covariance determinants and restricted trek separation in Gaussian graphical models , 2018 .

[14]  Marloes H. Maathuis,et al.  Structure Learning in Graphical Modeling , 2016, 1606.02359.

[15]  Aapo Hyvärinen,et al.  A Linear Non-Gaussian Acyclic Model for Causal Discovery , 2006, J. Mach. Learn. Res..

[16]  Judea Pearl,et al.  Equivalence and Synthesis of Causal Models , 1990, UAI.

[17]  Bernd Sturmfels,et al.  Algebraic geometry of Bayesian networks , 2005, J. Symb. Comput..

[18]  Pierre Comon,et al.  Handbook of Blind Source Separation: Independent Component Analysis and Applications , 2010 .

[19]  Lieven De Lathauwer,et al.  A Link between the Canonical Decomposition in Multilinear Algebra and Simultaneous Matrix Diagonalization , 2006, SIAM J. Matrix Anal. Appl..

[20]  Peter Buhlmann,et al.  Geometry of the faithfulness assumption in causal inference , 2012, 1207.0547.

[21]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[22]  I. Gessel,et al.  Binomial Determinants, Paths, and Hook Length Formulae , 1985 .

[23]  R. Evans,et al.  Constraints in Gaussian Graphical Models , 2019 .

[24]  Yury Polyanskiy,et al.  Algebraic methods of classifying directed graphical models , 2014, 2014 IEEE International Symposium on Information Theory.

[25]  Anima Anandkumar,et al.  Tensor decompositions for learning latent variable models , 2012, J. Mach. Learn. Res..

[26]  Aapo Hyvärinen,et al.  Pairwise likelihood ratios for estimation of non-Gaussian structural equation models , 2013, J. Mach. Learn. Res..

[27]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[28]  Patrik O. Hoyer,et al.  Discovering Cyclic Causal Models by Independent Components Analysis , 2008, UAI.

[29]  D. A. Kenny,et al.  Correlation and Causation. , 1982 .

[30]  T. Speed,et al.  Recursive causal models , 1984, Journal of the Australian Mathematical Society. Series A. Pure Mathematics and Statistics.

[31]  B. Lindström.  On the Vector Representations of Induced Matroids , 1973 .