CMStatistics 2016: Start Registration
View Submission - CFE
Title: Detecting Granger causality with a nonparametric information-based statistic Authors:  Hao Fang - University of Amsterdam (Netherlands) [presenting]
Cees Diks - University of Amsterdam (Netherlands)
Abstract: Testing causal effects has attracted much attention in the domains of many disciplines since Grangers pioneering work. The recent literature shows an increasing interest in testing for Granger non-causality in a general sense by nonparametric evaluation of conditional dependence. We introduce a novel nonparametric test based on the first order Taylor expansion of an information theoretic measure: transfer entropy. This new test statistic is shown to have an information-based interpretation for Granger non-causality. The proposed test avoids the impotence of the frequently-used test previously proposed as a result of the lack of the positive definiteness under some alternative circumstances. Asymptotic normality of the test statistic is achieved when all densities are estimated with appropriate sample-size dependent bandwidth, and practical guidelines for choosing bandwidth are formulated for specific cases. Simulation result confirms the usefulness of this test. Finally an application to financial data indicates the existence of bidirectional Granger causality.