Abstract
Recently independent component analysis (ICA) has been proposed for discovery of linear, non-Gaussian, and acyclic causal models (LiNGAM). As in practice the LiNGAM assumption usually does not exactly hold, in this paper we propose some methods to perform causality discovery even when LiNGAM is violated. The first method is ICA with a sparse separation matrix. By incorporating a suitable penalty term, the separation matrix produced by this method tends to satisfy the LiNGAM assumption. The other two methods are proposed to tackle nonlinearity in the data generation procedure, which violates the LiNGAM assumption. In the second method, the post-nonlinear mixing ICA model is exploited to do causality discovery when the nonlinearity is component-wise. The third method is proposed for the case where the nonlinear distortion in data generation is of arbitrary form, but smooth and weak. The separation system for such data is a linear transformation coupled with a nonlinear one, and the nonlinear one is as weak as possible such that it can be neglected when performing causality discovery. The linear causal relations in the data are then revealed. The proposed methods are applied to discover the causal relations in the Hong Kong stock market, and the last method works very well. The resulting causal diagram shows some interesting information in the stock market.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Almeida, L.B.: MISEP - linear and nonlinear ICA based on mutual information. Journal of Machine Learning Research 4, 1297–1318 (2003)
Dodge, Y., Rousson, V.: On asymmetric properties of the correlation coefficient in the regression setting. The American Statistician 55(1), 51–54 (2001)
Fan, J., Li, R.: Variable selection via nonconcave penalized likelihood and its oracle properties. Journal of the American Statist. Assoc. 96, 1348–1360 (2001)
Granovetter, M.: Business groups. In: Handbook of Economic Sociology, ch. 18, Princeton University Press, Princeton (1994)
Ho, R.Y., Strange, R., Piesse, J.: The structural and institutional features of the Hong Kong stock market: Implications for asset pricing. Research Paper 027, The Management Centre Research Papers, King’s College London (2004)
Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. John Wiley, Chichester (2001)
Jutten, C., Karhunen, J.: Advances in nonlinear blind source separation. In: Proc. ICA 2003, pp. 245–256 (2003); Invited paper in the special session on nonlinear ICA and BSS
Khanna, T., Rivkin, J.W.: Interorganizational ties and business group boundaries: Evidence from an emerging economy. Organization Science (forthcoming, 2006)
Pearl, J.: Causality: Models, Reasoning, and Inference. Cambridge University Press, Cambridge (2000)
Pham, D.T., Garat, P.: Blind separation of mixture of independent sources through a quasi-maximum likelihood approach. IEEE Trans. on Signal Processing 45(7), 1712–1725 (1997)
Shimizu, S., Hoyer, P.O., Hyvärinen, A., Kerminen, A.J.: A linear non-Gaussian acyclic model for causal discovery. Journal of Machine Learning Research (submitted, 2006)
Taleb, A., Jutten, C.: Source separation in post-nonlinear mixtures. IEEE Trans. on Signal Processing 47(10), 2807–2820 (1999)
Tibshirani, R.: Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society 58(1), 267–288 (1996)
Zhang, K., Chan, L.W.: Extended Gaussianization method for blind separation of post-nonlinear mixtures. Neural Computation 17(2), 425–452 (2005)
Zhang, K., Chan, L.W.: Nonlinear ICA with limited nonlinearity. Technical report, The Chinese Univerity of Hong Kong (2006) (to be available soon)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhang, K., Chan, LW. (2006). Extensions of ICA for Causality Discovery in the Hong Kong Stock Market. In: King, I., Wang, J., Chan, LW., Wang, D. (eds) Neural Information Processing. ICONIP 2006. Lecture Notes in Computer Science, vol 4234. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11893295_45
Download citation
DOI: https://doi.org/10.1007/11893295_45
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-46484-6
Online ISBN: 978-3-540-46485-3
eBook Packages: Computer ScienceComputer Science (R0)