Discovering Cyclic Causal Models by Independent Components Analysis

We generalize Shimizu et al's (2006) ICA-based approach for discovering linear non-Gaussian acyclic (LiNGAM) Structural Equation Models (SEMs) from causally sufficient, continuous-valued observational data. By relaxing the assumption that the generating SEM's graph is acyclic, we solve the more general problem of linear non-Gaussian (LiNG) SEM discovery. LiNG discovery algorithms output the distribution equivalence class of SEMs which, in the large sample limit, represents the population distribution. We apply a LiNG discovery algorithm to simulated data. Finally, we give sufficient conditions under which only one of the SEMs in the output class is 'stable'.

[1]  Thomas S. Richardson,et al.  A Discovery Algorithm for Directed Cyclic Graphs , 1996, UAI.

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

[3]  F. Fisher.  A Correspondence Principle for Simultaneous Equation Models , 1970 .

[4]  Denver Dash,et al.  Restructuring Dynamic Causal Systems in Equilibrium , 2005, AISTATS.

[5]  Peter Spirtes,et al.  Directed Cyclic Graphical Representations of Feedback Models , 1995, UAI.

[6]  Geoffrey James Wyatt.  Macroeconomic Models in a Causal Framework , 2004 .

[7]  Thomas S. Richardson,et al.  A Polynomial-Time Algorithm for Deciding Markov Equivalence of Directed Cyclic Graphical Models , 1996, UAI 1996.

[8]  Lai-Wan Chan,et al.  ICA with Sparse Connections , 2006, IDEAL.

[9]  Harold W. Kuhn,et al.  The Hungarian method for the assignment problem , 1955, 50 Years of Integer Programming.

[11]  Patrik O. Hoyer,et al.  Estimation of linear, non-gaussian causal models in the presence of confounding latent variables , 2006, Probabilistic Graphical Models.

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

[13]  Chandra R. Chegireddy,et al.  Algorithms for finding K-best perfect matchings , 1987, Discret. Appl. Math..

[14]  E. Oja,et al.  Independent Component Analysis , 2013 .

[15]  J. Pearl.  Causal diagrams for empirical research , 1995 .

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

[17]  Pierre Comon.  Independent component analysis - a new concept? signal processing , 1994 .

[18]  Robert H. Strotz,et al.  Recursive versus non-recursive systems: An attempt at a synthesis , 2017 .

[19]  Aapo Hyvärinen,et al.  Causal discovery of linear acyclic models with arbitrary distributions , 2008, UAI.