Causal Discovery using Model Invariance through Knockoff Interventions

TL;DR

This paper introduces a novel causality inference method for nonlinear multivariate time series by leveraging model invariance through Knockoff interventions, outperforming existing techniques.

Contribution

It presents a new approach combining DeepAR and Knockoff interventions to identify causal predictors via model invariance testing in time series.

Findings

Outperforms VAR Granger causality, VARLiNGAM, and PCMCI+

Effective on both real and synthetic data

Demonstrates robustness in nonlinear settings

Abstract

Cause-effect analysis is crucial to understand the underlying mechanism of a system. We propose to exploit model invariance through interventions on the predictors to infer causality in nonlinear multivariate systems of time series. We model nonlinear interactions in time series using DeepAR and then expose the model to different environments using Knockoffs-based interventions to test model invariance. Knockoff samples are pairwise exchangeable, in-distribution and statistically null variables generated without knowing the response. We test model invariance where we show that the distribution of the response residual does not change significantly upon interventions on non-causal predictors. We evaluate our method on real and synthetically generated time series. Overall our method outperforms other widely used causality methods, i.e, VAR Granger causality, VARLiNGAM and PCMCI+.