Abstract
In this chapter, we will discuss the problem of estimating the network graphs of state-space representations based on observed data: we observe the output generated by each node of a network of state-space representations, and we would like to reconstruct the communication graph of the network, i.e., we would like to find out which nodes exchange information. The potential exchange of information is not assumed to be observable, i.e., it may take place via hidden internal states. We present an approach based on the notion of Granger causality. The essence of this approach is that there exists a communication link between two nodes, if the outputs generated by the corresponding nodes are related by Granger causality. More precisely, we show an equivalence between the existence of state-space representation in which subsystems corresponding to certain nodes exchange information, and the presence of Granger causality relation between the outputs generated by those subsystems. Since Granger causality can be checked based on observed data, these results open up the possibility of data-driven reverse engineering of the communication graph. We will discuss the case of stochastic linear time-invariant systems, and then the case of stochastic bilinear/LPV/switched systems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Stationarity implies that the (co)variance matrices are time-independent.
- 2.
For any other symmetric solution \(\tilde{\Sigma }\), the matrix \(\tilde{\Sigma }\!-\!\Sigma \) is positive definite.
- 3.
See [34, Definition 9.4.1] for the definition of coercivity.
References
Amblard, P.-O., Michel, O.J.: On directed information theory and granger causality graphs. J. Comput. Neurosci. 30(1), 7–16 (2011)
Barnett, L., Barrett, A.B., Seth, A.K.: Granger causality and transfer entropy are equivalent for Gaussian variables. Phys. Rev. Lett. 103, 238701 (2009)
Bolognani, S., Bof, N., Michelotti, D., Muraro, R., Schenato, L.: Identification of power distribution network topology via voltage correlation analysis. In: 52nd IEEE Conference on Decision and Control, pp. 1659–1664 (2013)
Bombois, X., Colin, K., Van Den Hof, P.M.J., Hjalmarsson, H.: On the informativity of direct identification experiments in dynamical networks. Working paper or preprint, August 2021
Caines, P.E.: Weak and strong feedback free processes. IEEE Trans. Autom. Control 21(5), 737–739 (1976)
Caines, P.E., Chan, C.: Feedback between stationary stochastic processes. IEEE Trans. Autom. Control 20(4), 498–508 (1975)
Caines, P.E., Deardon, R., Wynn, H.P.: Conditional orthogonality and conditional stochastic realization. In: Rantzer, A., Byrnes, C.I. (eds.) Directions in Mathematical Systems Theory and Optimization, vol. 286, pp. 71–84. Springer, Berlin, Heidelberg (2003)
Caines, P.E., Deardon, R., Wynn, H.P.: Bayes nets of time series: stochastic realizations and projections. In: Pronzato, L., Zhigljavsky, A. (eds.) Optimal Design and Related Areas in Optimization and Statistics. Springer Optimization and its Applications, vol. 28, pp. 155–166. Springer, New York (2009)
Caines, P.E.: Linear Stochastic Systems. John Wiley and Sons, New-York (1998)
Dankers, A.G.: System identification in dynamic networks. Ph.D. thesis, Delft University of Technology (2014)
David, O.: fMRI connectivity, meaning and empiricism: comments on: Roebroeck et al. the identification of interacting networks in the brain using fMRI: model selection, causality and deconvolution. NeuroImage 58(2), 306–309 (2011)
Eichler, M.: Granger causality and path diagrams for multivariate time series. J. Econ. 137(2), 334–353 (2007)
Eichler, M.: Graphical modelling of multivariate time series. Probab. Theory Relat. Fields 153(1), 233–268 (2012)
Friston, K.J., Harrison, L., Penny, W.: Dynamic causal modeling. NeuroImage 19(4), 1273–1302 (2003)
Gevers, M.R., Anderson, B.: On jointly stationary feedback-free stochastic processes. IEEE Trans. Autom. Control 27(2), 431–436 (1982)
Goebel, R., Roebroeck, A., Kim, D.-S., Formisano, E.: Investigating directed cortical interactions in time-resolved fMRI data using vector autoregressive modeling and Granger causality mapping. Magn. Reson. Imaging 21, 1251–1261 (2003)
Gonçalves, J., Howes, R., Warnick, S.: Dynamical structure functions for the reverse engineering of LTI networks. In: 46th IEEE Conference on Decision and Control, pp. 1516–1522 (2007)
Granger, C.W.J.: Economic processes involving feedback. Inf. Control 6(1), 28–48 (1963)
Havlicek, M., Roebroeck, A., Friston, K., Gardumi, A., Ivanov, D., Uludag, K.: Physiologically informed dynamic causal modeling of fMRI data. NeuroImage 122, 355–372 (2015)
Howes, R., Eccleston, L., Gonçalves, J., Stan, G.-B., Warnick, S.: Dynamical structure analysis of sparsity and minimality heuristics for reconstruction of biochemical networks. In: 47th IEEE Conference on Decision and Control (2008)
Hsiao, C.: Autoregressive modelling and causal ordering of econometric variables. J. Econ. Dyn. Control 4, 243–259 (1982)
Isidori, A.: Direct construction of minimal bilinear realizations from nonlinear input-output maps. IEEE Trans. Autom. Control 626–631 (1973)
Jozsa, M.: Relationship between Granger non-causality and network graphs of state-space representations. Ph.D. thesis, University of Groningen (2019)
Jozsa, M., Petreczky, M., Camlibel, M.K.: Causality based graph structure of stochastic linear state-space representations. In: 56th IEEE Conference on Decision and Control, pp. 2442–2447 (2017)
Jozsa, M., Petreczky, M., Camlibel, M.K.: Relationship between causality of stochastic processes and zero blocks of their joint innovation transfer matrices. In: 20th World Congress of the International Federation of Automatic Control, pp. 4954–4959 (2017)
Jozsa, M., Petreczky, M., Camlibel, M.K.: Relationship between Granger non-causality and network graph of state-space representations. IEEE Trans. Autom. Control 64(8), 912–927 (2019)
Jozsa, M., Petreczky, M., Camlibel, M.K.: Causality and network graph in general bilinear state-space representations. IEEE Trans. Autom. Control 65(8), 3623–3630 (2020)
Julius, A.A., Zavlanos, M., Boyd, S., Pappas, G.J.: Genetic network identification using convex programming. IET Syst. Biol. 3, 155–166 (2009)
Kang, T., Moore, R., Li, Y., Sontag, E.D., Bleris, L.: Discriminating direct and indirect connectivities in biological networks. Natl. Acad. Sci. USA 112, 12893–12898 (2015)
Katayama, T.: Subspace Methods for System Identification. Springer-Verlag (2005)
Kempker, P.L.: Coordination control of linear systems. Ph.D. thesis, Amsterdam: Vrije Universiteit (2012)
Kempker, P.L., Ran, A.C.M., van Schuppen, J.H.: Construction and minimality of coordinated linear systems. Linear Algebr. Appl. 452, 202–236 (2014)
Kramer, G.: Directed information for channels with feedback. Ph.D. thesis, Swiss Federal Institute of Technology Zürich (1998)
Lindquist, A., Picci, G.: Linear Stochastic Systems: a Geometric Approach to Modeling. Estimation and Identification, Springe, Berlin (2015)
Massey, J.L.: Causality, feedback and directed information. In: International Symposium on Information Theory and its Applications, pp. 27–30 (1990)
Nordling, T.E.M., Jacobsen, E.W.: On sparsity as a criterion in reconstructing biochemical networks. In: 18th IFAC World Congress, pp. 11672–11678 (2011)
Pambakian, N.: LQG coordination control. Master’s thesis, Delft University of Technology (2011)
Pearl, J.: Causality: Models, Reasoning and Inference, 1st edn. Cambridge University Press (2000)
Penny, W.D., Stephan, K.E., Mechelli, A., Friston, K.J.: Comparing dynamic causal models. NeuroImage 22(3), 1157–1172 (2004)
Petreczky, M., Vidal, R.: Realization theory for a class of stochastic bilinear systems. IEEE Trans. Autom. Control 63(1), 69–84 (2017)
Ran, A.C.M., van Schuppen, J.H.: Coordinated linear systems. In: Coordination Control of Distributed Systems. Lecture Notes in Control and Information Sciences, vol. 456, pp. 113–121 (2014)
Roebroeck, A., Formisano, E., Goebel, R.: Reply to Friston and David: after comments on: the identification of interacting networks in the brain using fMRI: model selection, causality and deconvolution. NeuroImage 58(2), 310–311 (2011)
Roebroeck, A., Seth, A.K., Valdes-Sosa, P.A.: Causal time series analysis of functional magnetic resonance imaging data. J. Mach. Learn. Res. Proc. Track 12, 65–94 (2011)
Valdes-Sosa, P.A., Roebroeck, A., Daunizeau, J., Friston, K.J.: Effective connectivity: influence, causality and biophysical modeling. NeuroImage 58(2), 339–361 (2011)
Van den Hof, P.M.J., Dankers, A., Heuberger, P.S.C., Bombois, X.: Identification of dynamic models in complex networks with prediction error methods-basic methods for consistent module estimates. Automatica 49(10), 2994–3006 (2013)
Weerts, H.H.M.: Identifiability and identification methods for dynamic networks. Ph.D. thesis, Eindhoven University of Technology (2018)
Weerts, H.H.M., Van den Hof, P.M.J., Dankers, A.G.: Identifiability of linear dynamic networks. Automatica 89, 247–258 (2018)
Yuan, Y., Glover, K., Gonçalves, J.: On minimal realisations of dynamical structure functions. Automatica 55, 159–164 (2015)
Zhang, W., Liu, W., Zang, C., Liu, L.: Multi-agent system based integrated solution for topology identification and state estimation. IEEE Trans. Ind. Inform. 13(2), 714–724 (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Józsa, M., Petreczky, M., Camlibel, M.K. (2024). Relating the Network Graphs of State-Space Representations to Granger Causality Conditions. In: Postoyan, R., Frasca, P., Panteley, E., Zaccarian, L. (eds) Hybrid and Networked Dynamical Systems. Lecture Notes in Control and Information Sciences, vol 493. Springer, Cham. https://doi.org/10.1007/978-3-031-49555-7_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-49555-7_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-49554-0
Online ISBN: 978-3-031-49555-7
eBook Packages: EngineeringEngineering (R0)