A Kernel Test for Causal Association via Noise Contrastive Backdoor Adjustment

2021/11/25 by Hu, Robert, Sejdinovic, Dino, Evans, Robin J. · 2 citations
#FOS: Computer and information sciences #Machine Learning (stat.ML) #Methodology (stat.ME)

paper · doi:10.48550/arxiv.2111.13226

Abstract

Causal inference grows increasingly complex as the number of confounders increases. Given treatments X, confounders Z and outcomes Y, we develop a non-parametric method to test the do-null hypothesis H₀: p(y|\it do(X=x))=p(y) against the general alternative. Building on the Hilbert Schmidt Independence Criterion (HSIC) for marginal independence testing, we propose backdoor-HSIC (bd-HSIC) and demonstrate that it is calibrated and has power for both binary and continuous treatments under a large number of confounders. Additionally, we establish convergence properties of the estimators of covariance operators used in bd-HSIC. We investigate the advantages and disadvantages of bd-HSIC against parametric tests as well as the importance of using the do-null testing in contrast to marginal independence testing or conditional independence testing. A complete implementation can be found at \hyperlinkhttps://github.com/MrHuff/kgformulahttps://github.com/MrHuff/kgformula.

Cited by

• RI-Loss: A Learnable Residual-Informed Loss for Time Series Forecasting
• ActiveCQ: Active Estimation of Causal Quantities

Related

• Kernel-based Tests for Joint Independence
• Nonparametric Independence Testing for Small Sample Sizes
• A Permutation-Free Kernel Independence Test
• A Scalable Conditional Independence Test for Nonlinear, Non-Gaussian Data
• A fast and accurate kernel-based independence test with applications to high-dimensional and functional data