DIET: Conditional independence testing with marginal dependence measures of residual information

TL;DR

DIET introduces a novel, computationally efficient method for conditional independence testing that leverages marginal dependence measures, outperforming existing techniques in power and accuracy.

Contribution

The paper proposes DIET, a new algorithm that tests conditional independence using marginal dependence measures, avoiding costly model fitting and maintaining statistical power.

Findings

DIET achieves higher power than existing CRT methods on benchmarks.

DIET maintains finite sample type-1 error control under certain conditions.

Using mutual information as a test statistic makes DIET the most powerful among tractable tests.

Abstract

Conditional randomization tests (CRTs) assess whether a variable $x$ is predictive of another variable $y$, having observed covariates $z$. CRTs require fitting a large number of predictive models, which is often computationally intractable. Existing solutions to reduce the cost of CRTs typically split the dataset into a train and test portion, or rely on heuristics for interactions, both of which lead to a loss in power. We propose the decoupled independence test (DIET), an algorithm that avoids both of these issues by leveraging marginal independence statistics to test conditional independence relationships. DIET tests the marginal independence of two random variables: $F(x \mid z)$ and $F(y \mid z)$ where $F(\cdot \mid z)$ is a conditional cumulative distribution function (CDF). These variables are termed "information residuals." We give sufficient conditions for DIET to achieve finite sample type-1 error control and power greater than the type-1 error rate. We then prove that when using the mutual information between the information residuals as a test statistic, DIET yields the most powerful conditionally valid test. Finally, we show DIET achieves higher power than other tractable CRTs on several synthetic and real benchmarks.