A Heuristic Partial-Correlation-Based Algorithm for Causal Relationship Discovery on Continuous Data
In this paper, we propose a heuristic partial-correlation- based (HP) algorithm to discover causal structures of Bayesian networks with continuous variables. There are two advantages of HP algorithm compared with existing ones: the first is that HP algorithm has a polynomial time complexity in the worst case, and the second HP algorithm can be applied to the data generated by linear simultaneous equation model, without assuming data following multivariate Gaussian distribution. Empirical results show that HP algorithm outperforms existing algorithms in both accuracy and run time.
KeywordsBayesian Network Partial Correlation Causal Structure Multivariate Gaussian Distribution Conditional Correlation
Unable to display preview. Download preview PDF.
- 1.Pearl, J., Verma, T.: A Theory of Inferred Causation. In: Proc. of the Second Int. Conf. on Principles of Knowledge Representation and Reasoning. Morgan Kaufmann, San Francisco (1991)Google Scholar
- 3.Margaritis, D., Thrun, S.: A Bayesian Multiresolution Independence Test for Continuous Variables. In: Proc. of the 17th Conf. in Uncertainty in Artificial Intelligence, San Francisco, pp. 346–353 (2001)Google Scholar
- 4.Margaritis, D.: Distribution-Free Learning of Bayesian Network Structure in Continuous Domains. In: Proc. of the 20th National Conf. on AI (2005)Google Scholar
- 9.Opgen-Rhein, R., Strimmer, K.: From Correlation to Causation Networks: A Simple Approximate Learning Algorithm and Its Application to High-Dimensional Plant Gene Expression Data. In: BMC Systems Biology (2007)Google Scholar
- 10.Bayesian Network Repository, http://compbio.cs.huji.ac.il/Repository/
- 11.Scheines, R., Spirtes, P., Glymour, C., Meek, C., Richardson, T.: The Tetrad Project: Constraint Based Aids to Causal Model Specification. Technical Report, Carnegie Mellon University, Dpt. of Philosophy (1995)Google Scholar