Sparse Causality Network Retrieval from Short Time Series

We investigate how efficiently a known underlying sparse causality structure of a simulated multivariate linear process can be retrieved from the analysis of time series of short lengths. Causality is quantified from conditional transfer entropy and the network is constructed by retaining only the statistically validated contributions. We compare results from three methodologies: two commonly used regularization methods, Glasso and ridge, and a newly introduced technique, LoGo, based on the combination of information filtering network and graphical modelling. For these three methodologies we explore the regions of time series lengths and model-parameters where a significant fraction of true causality links is retrieved. We conclude that when time series are short, with their lengths shorter than the number of variables, sparse models are better suited to uncover true causality links with LoGo retrieving the true causality network more accurately than Glasso and ridge.

[1]  Michael Elad,et al.  From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images , 2009, SIAM Rev..

[2]  S. S. Wilks.  The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses , 1938 .

[3]  Imre Kondor,et al.  Replica approach to mean-variance portfolio optimization , 2016, 1606.08679.

[4]  Guido Previde Massara,et al.  Network Filtering for Big Data: Triangulated Maximally Filtered Graph , 2015, J. Complex Networks.

[5]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[6]  Wolfram Barfuss,et al.  Parsimonious modeling with Information Filtering Networks , 2016, Physical review. E.

[7]  R. Tibshirani,et al.  Covariance‐regularized regression and classification for high dimensional problems , 2009, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[8]  Schreiber,et al.  Measuring information transfer , 2000, Physical review letters.

[9]  James D. Hamilton.  Time Series Analysis , 1994 .

[10]  Gerhard Kramer,et al.  Directed information for channels with feedback , 1998 .

[11]  A. E. Hoerl,et al.  Ridge regression: biased estimation for nonorthogonal problems , 2000 .

[12]  J. Kruskal.  On the shortest spanning subtree of a graph and the traveling salesman problem , 1956 .

[13]  I. Kondor,et al.  Fluctuation-bias trade-off in portfolio optimization under Expected Shortfall with $\ell_2$ regularization , 2016 .

[14]  J. Massey.  CAUSALITY, FEEDBACK AND DIRECTED INFORMATION , 1990 .

[15]  N. Giri.  Multivariate Statistical Analysis : Revised And Expanded , 2003 .

[16]  M Tumminello,et al.  A tool for filtering information in complex systems. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[17]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[18]  Marvin H. J. Gruber.  Improving Efficiency by Shrinkage: The James--Stein and Ridge Regression Estimators , 1998 .

[19]  Olivier J. J. Michel,et al.  The relation between Granger causality and directed information theory: a review , 2012, Entropy.

[20]  R. Mantegna.  Hierarchical structure in financial markets , 1998, cond-mat/9802256.

[21]  N. Meinshausen,et al.  High-dimensional graphs and variable selection with the Lasso , 2006, math/0608017.

[22]  Tomaso Aste,et al.  Measures of Causality in Complex Datasets with Application to Financial Data , 2014, Entropy.

[23]  Hengqing Tong,et al.  Multivariate Statistical Analysis and Data Analysis , 2011 .

[24]  John A. Swets,et al.  Signal Detection Theory and ROC Analysis in Psychology and Diagnostics: Collected Papers , 1996 .

[25]  A Tikhonov,et al.  Solution of Incorrectly Formulated Problems and the Regularization Method , 1963 .

[26]  T. Di Matteo,et al.  Complex networks on hyperbolic surfaces , 2004, cond-mat/0408443.

[27]  George Sugihara,et al.  Spatial convergent cross mapping to detect causal relationships from short time series. , 2015, Ecology.

[28]  R. Tibshirani,et al.  Sparse inverse covariance estimation with the graphical lasso. , 2008, Biostatistics.

[29]  Q. Vuong.  Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses , 1989 .

[30]  George Sugihara,et al.  Detecting Causality in Complex Ecosystems , 2012, Science.

[31]  Sergios Theodoridis,et al.  Sparsity-Aware Learning and Compressed Sensing: An Overview , 2012, ArXiv.