Advertisement

Neural Causality Detection for Multi-dimensional Point Processes

  • Tianyu Wang
  • Christian Walder
  • Tom Gedeon
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11304)

Abstract

In the big data era, while correlation detection is relatively straightforward and successfully addressed by many techniques, causality detection does not have a generally-used solution. Causality provides valuable insights into data and guides further studies. With the overall assumption that causal influence can only be from prior history events, time plays an essential part in causality analysis, and this important feature means the data with strict temporal structure needs to be modelled. Traditionally, temporal point processes are employed to model data containing temporal structure information. The heuristic parameterization property of such models makes the task difficult. Domain related knowledge are needed to design proper parameterization. Recently, Recurrent Neural Networks (RNNs) have been used for time-related data modelling. RNN’s trainable parameterization considerably reduces the dependency on domain-related knowledge. In this work, we show that combining neural network techniques with Granger causality framework has great potential by presenting an RNN model integrated with a Granger causality framework. The experimental results show that the same network structure can be applied to a variety of datasets and causalities are detected successfully.

Keywords

Granger causality Recurrent neural network Temporal point process 

References

  1. 1.
    Ancona, N., Marinazzo, D., Stramaglia, S.: Radial basis function approach to nonlinear granger causality of time series. Phys. Rev. E Stat. Nonlinear Soft Matter Phys. 70(5 Pt 2), 056221 (2004)Google Scholar
  2. 2.
    Bacry, E., Delattre, S., Hoffmann, M., Muzy, J.: Some limit theorems for Hawkes processes and application to financial statistics. Stoch. Process. Their Appl. 123(7), 2475–2499 (2013).  https://doi.org/10.1016/j.spa.2013.04.007. http://www.sciencedirect.com/science/article/pii/S0304414913001026. A Special Issue on the Occasion of the 2013 International Year of StatisticsMathSciNetCrossRefGoogle Scholar
  3. 3.
    Chen, Y., Rangarajan, G., Feng, J., Ding, M.: Analyzing multiple nonlinear time series with extended Granger causality. Phys. Lett. A 324, 26–35 (2004).  https://doi.org/10.1016/j.physleta.2004.02.032MathSciNetCrossRefzbMATHGoogle Scholar
  4. 4.
    Du, N., Dai, H., Trivedi, R., Upadhyay, U., Gomez-Rodriguez, M., Song, L.: Recurrent marked temporal point processes: embedding event history to vector. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016, pp. 1555–1564. ACM, New York (2016).  https://doi.org/10.1145/2939672.2939875. http://doi.acm.org/10.1145/2939672.2939875
  5. 5.
    Freiwald, W.A., et al.: Testing non-linearity and directedness of interactions between neural groups in the macaque inferotemporal cortex. J. Neurosci. Methods 94(1), 105–119 (1999).  https://doi.org/10.1016/S0165-0270(99)00129-6. http://www.sciencedirect.com/science/article/pii/S0165027099001296CrossRefGoogle Scholar
  6. 6.
    Granger, C.W.J.: Investigating causal relations by econometric models and cross-spectral methods. Econometrica 37(3), 424–438 (1969). http://www.jstor.org/stable/1912791CrossRefGoogle Scholar
  7. 7.
    Guttorp, P.: An introduction to the theory of point processes (D. J. Daley and D. Vere-Jones). SIAM Rev. 32(1), 175-2 (1990). https://search.proquest.com/docview/926135796?accountid=8330. Copyright \(\copyright \) 1990 Society for Industrial and Applied Mathematics. Last updated 05 Mar 2012. CODEN - SIREAD
  8. 8.
    Kim, S., Putrino, D., Ghosh, S., Brown, E.N.: A granger causality measure for point process models of ensemble neural spiking activity. PLOS Comput. Biol. 7(3), 1–13 (2011).  https://doi.org/10.1371/journal.pcbi.1001110MathSciNetCrossRefGoogle Scholar
  9. 9.
    Liniger, T.J.: Multivariate Hawkes processes. Ph.D. thesis, ETH Zurich. Diss., Eidgenössische Technische Hochschule ETH Zürich, Nr. 18403 (2009).  https://doi.org/10.3929/ethz-a-006037599
  10. 10.
    Marinazzo, D., Pellicoro, M., Stramaglia, S.: Nonlinear parametric model for Granger causality of time series 73(6), 066216 (2006).  https://doi.org/10.1103/PhysRevE.73.066216CrossRefGoogle Scholar
  11. 11.
    Marinazzo, D., Pellicoro, M., Stramaglia, S.: Kernel method for nonlinear granger causality. Phys. Rev. Lett. 100(14), 144103 (2008).  https://doi.org/10.1103/PhysRevLett.100.144103
  12. 12.
    Ogata, Y.: Statistical models for earthquake occurrences and residual analysis for point processes. J. Am. Stat. Assoc. 83(401), 9–27 (1988).  https://doi.org/10.1080/01621459.1988.10478560. https://amstat.tandfonline.com/doi/abs/10.1080/01621459.1988.10478560CrossRefGoogle Scholar
  13. 13.
    Toke, I.M.: An introduction to Hawkes processes with applications to finance, p. 6, February 2011Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Australia National University College of Engineering and Computer ScienceCanberraAustralia

Personalised recommendations