Learning Granger causality graphs for multivariate nonlinear time series

Abstract

An information theory method is proposed to test the Granger causality and contemporaneous conditional independence in Granger causality graph models. In the graphs, the vertex set denotes the component series of the multivariate time series, and the directed edges denote causal dependence, while the undirected edges reflect the instantaneous dependence. The presence of the edges is measured by a statistics based on conditional mutual information and tested by a permutation procedure. Furthermore, for the existed relations, a statistics based on the difference between general conditional mutual information and linear conditional mutual information is proposed to test the nonlinearity. The significance of the nonlinear test statistics is determined by a bootstrap method based on surrogate data. We investigate the finite sample behavior of the procedure through simulation time series with different dependence structures, including linear and nonlinear relations.

Keywords

Multivariate nonlinear time series / Granger causality graph / conditional mutual information / surrogate data

References

[1] Ancona N., Marinazzo D., Stramaglia S.. Radial basis function approach to nonlinear Granger causality of time series. Physical Review E, 2004, 70(5): 056221.

[2] Brillinger D.R.. Remarks concerning graphical models for time series and point processes. Revista de Econometria, 1996, 16: 1-23.

[3] Chavez M., Martinerie J., Quyen M.L.V.. Statistical assessment of nonlinear causality: application to epileptic EEG signals. Journal of Neuroscience Methods, 2003, 124: 113-128.

[4] Chen Y.H., Rangarajan G., Feng J.F., Ding M.Z.. Analyzing multiple nonlinear time series with extended Granger causality. Physics Letters A, 2004, 324: 26-35.

[5] Dahlhaus, R. (2000). Graphical interaction models for multivariate time series. Metrika, 157–172

[6] Diks C., DeGoede J.. Broer H.W., Krauskopf B., Vegter G.. A general nonparametric bootstrap test for Granger causality. Global Analysis of Dynamical Systems, 2001, Bristol: IoP Publishing 391-403.

[7] Eichler, M. (2001). Graphical modelling of multivariate time series. To appear in Probability Theory and Related Fields

[8] Gao, W. & Tian, Z. (2008). Detecting lags in nonlinear models using general mutual information. To appear in Journal of Mathematical Research and Exposition

[9] Granger C.W.J.. Investigating causal relations by econometric models and cross-spectral methods. Econometrica, 1969, 37: 424-438.

[10] Grassberger P.. Finite sample correction to entropy and dimension estimates. Physics Letters A, 1988, 128: 369-373.

[11] Hoover K.D.. Causality in Macroeconomics, 2001, Cambridge: Cambridge University Press

[12] Hsaio C.. Autoregressive modelling and causal ordering of economic variables. Journal of Economic Dynamic and Control, 1982, 4: 243-259.

[13] Kaminski M., Ding M., Truccolo W.A., Bressler S.L.. Evaluating causal relations in neural systems: Granger causality, directed transfer function and statistical assessment of significance. Biological Cybernetics, 2001, 85: 145-157.

[14] Lauritzen S.L.. Graphical Models, 1996, Oxford: Oxford University Press

[15] Marinazzo D., Pellicoro M., Stramaglia S.. Kernel method for nonlinear Granger causality. Physics Letters A, 2008, 100: 144103.

[16] Marinazzo D., Pellicoro M., Stramaglia S.. Kernel Granger causality and the analysis of dynamical networks. Physical Review E, 2008, 77: 056215.

[17] Palus M.. Detecting nonlinearity in multivariate time series. Physics Letters A, 1996, 213: 138-147.

[18] Palus M., Albrecht V., Dvorak I.. Information-theoretic test for nonlinearity in time series. Physics Letters A, 1993, 175: 203-209.

[19] Prichard D., Theiler J.. Generalized redundancies for time series analysis. Physica D, 1995, 84: 476-453.

[20] Prichard D., Theiler J.. Generating surrogate data for time series with several simultaneously measured variables. Physical Review Letters, 1994, 73: 951-954.

[21] Rapp P., Celucci C.J., Watanabe T.A., Albano A.M., Xchmah T.I.. Surrogate data pathologies and the falsepositive rejection of the null hypothesis. International Journal of Bifurcation and Chaos, 2001, 11(4): 983-987.

[22] Reale M., Wilson G.T.. The sampling properties of conditional independence graphs for structural vector autoregressions. Biometrika, 2002, 89: 457-461.

[23] Whittaker J.. Graphical Models in Applied Multivariate Statistics, 1990, Chichester: John Wiley.