The Generalised Kernel Covariance Measure

TL;DR

The paper introduces GKCM, a flexible kernel-based conditional independence test that improves computational efficiency and calibration over existing methods, with strong theoretical guarantees and superior empirical performance.

Contribution

It proposes GKCM, a regression-model-agnostic kernel CI test that broadens applicability and enhances performance compared to existing kernel-based tests.

Findings

GKCM outperforms state-of-the-art CI tests in simulations.

GKCM achieves better type I error control.

GKCM maintains competitive or superior power across diverse data-generating processes.

Abstract

We consider the problem of conditional independence (CI) testing and adopt a kernel-based approach. Kernel-based CI tests embed variables in reproducing kernel Hilbert spaces, regress their embeddings on the conditioning variables, and test the resulting residuals for marginal independence. This approach yields tests that are sensitive to a broad range of conditional dependencies. Existing methods, however, rely heavily on kernel ridge regression, which is computationally expensive when properly tuned and yields poorly calibrated tests when left untuned, which limits their practical usefulness. We propose the Generalised Kernel Covariance Measure (GKCM), a regression-model-agnostic kernel-based CI test that accommodates a broad class of regression estimators. Building on the Generalised Hilbertian Covariance Measure framework (Lundborg et al., 2022), we characterise conditions under which GKCM satisfies uniform asymptotic level guarantees. In simulations, GKCM paired with tree-based regression models frequently outperforms state-of-the-art CI tests across a diverse range of data-generating processes, achieving better type I error control and competitive or superior power.