6533b82cfe1ef96bd129006d
RESEARCH PRODUCT
Explicit Granger causality in kernel Hilbert spaces
Gustau Camps-vallsMaria PilesDiego Buesosubject
Series (mathematics)Dynamical systems theoryHilbert spaceFOS: Physical sciencesNonlinear Sciences - Chaotic Dynamics01 natural sciences010305 fluids & plasmassymbols.namesakeKernel methodGranger causalityPhysics - Data Analysis Statistics and ProbabilityKernel (statistics)Arrow of time0103 physical sciencesRademacher complexitysymbolsApplied mathematicsChaotic Dynamics (nlin.CD)010306 general physicsData Analysis Statistics and Probability (physics.data-an)Mathematicsdescription
Granger causality (GC) is undoubtedly the most widely used method to infer cause-effect relations from observational time series. Several nonlinear alternatives to GC have been proposed based on kernel methods. We generalize kernel Granger causality by considering the variables cross-relations explicitly in Hilbert spaces. The framework is shown to generalize the linear and kernel GC methods, and comes with tighter bounds of performance based on Rademacher complexity. We successfully evaluate its performance in standard dynamical systems, as well as to identify the arrow of time in coupled R\"ossler systems, and is exploited to disclose the El Ni\~no-Southern Oscillation (ENSO) phenomenon footprints on soil moisture globally.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2020-05-14 |