A Kernel Independence Test for Random Processes

Abstract

A new non parametric approach to the problem of testing the independence of two random process is developed. The test statistic is the Hilbert Schmidt Independence Criterion (HSIC), which was used previously in testing independence for i.i.d pairs of variables. The asymptotic behaviour of HSIC is established when computed from samples drawn from random processes. It is shown that earlier bootstrap procedures which worked in the i.i.d. case will fail for random processes, and an alternative consistent estimate of the p-values is proposed. Tests on artificial data and real-world Forex data indicate that the new test procedure discovers dependence which is missed by linear approaches, while the earlier bootstrap procedure returns an elevated number of false positives. The code is available online: https://github.com/kacperChwialkowski/HSIC .

In Proceedings of The 31st International Conference on Machine Learning

References in corpus (1)

• Brownian distance covariance

Cited by in corpus (19)

• Kernel Mean Embedding of Distributions: A Review and Beyond
• Learning Decentralized Controllers for Robot Swarms with Graph Neural Networks
• Large-Scale Kernel Methods for Independence Testing
• Learning Granger Causality for Hawkes Processes
• Measuring Association on Topological Spaces Using Kernels and Geometric Graphs
• A Wild Bootstrap for Degenerate Kernel Tests
• Does Terrorism Trigger Online Hate Speech? On the Association of Events and Time Series
• The Arrow of Time in Multivariate Time Series
• Causal Inference from Slowly Varying Nonstationary Processes
• Nonparametric Independence Testing for Small Sample Sizes
• On Empirical Risk Minimization with Dependent and Heavy-Tailed Data
• A Kernel Test for Three-Variable Interactions with Random Processes
• Robust Independence Tests with Finite Sample Guarantees for Synchronous Stochastic Linear Systems
• Identifying Causal Structure in Dynamical Systems
• A kernel test for quasi-independence
• Intrinsic Sliced Wasserstein Distances for Comparing Collections of Probability Distributions on Manifolds and Graphs
• Metric on random dynamical systems with vector-valued reproducing kernel Hilbert spaces
• Estimating Rényi's $\alpha$-Cross-Entropies in a Matrix-Based Way
• A Kernel Two-sample Test for Dynamical Systems