Intervention-Based Time Series Causal Discovery via Simulator-Generated Interventional Distributions

TL;DR

This paper introduces SVAR-FM, a causal discovery framework for time series that uses physics-based simulation and flow matching to identify causal effects, with theoretical guarantees and empirical validation.

Contribution

The paper presents a novel method combining simulator-based interventions with flow matching, providing identifiability and error bounds for causal discovery in time series.

Findings

SVAR-FM recovers correct causal signs where observational methods fail due to confounding.

Theoretical proof of identifiability under coverage conditions.

Empirical validation across four scientific domains and a physics case study.

Abstract

We propose SVAR-FM (Structural VAR with Flow Matching), a framework for time series causal discovery that treats a physics-based simulator as a mechanical realization of Pearl's do operator. Clamping a variable inside the simulator physically severs confounding paths, producing interventional data by construction. Conditional Flow Matching then learns the nonlinear interventional conditionals. Theoretically, we prove that the full structural VAR becomes identifiable under a coverage condition on the simulator-clampable variables, and derive an end-to-end error bound that decomposes into Monte Carlo, simulator fidelity, and Flow Matching terms. A sign-flip corollary predicts that when simulator accuracy falls below a threshold, the estimated causal effect reverses sign. Empirically, a benchmark across four scientific domains confirms that SVAR-FM recovers the correct causal sign where observational methods produce sign-reversed estimates due to confounding. A case study in ultrafast laser physics verifies the sign-flip prediction by physically varying the accuracy level of a first-principles quantum solver: the low-accuracy setting reverses the causal sign, while the high-accuracy setting recovers the correct direction (R-squared = 0.983, zero bias).