TCD-Arena: Assessing Robustness of Time Series Causal Discovery Methods Against Assumption Violations

2026-05-04

AI Summary

Existing methods for causal discovery in time series rely on strong assumptions yet lack systematic evaluation of their robustness under assumption violations. To address this gap, this work proposes TCD-Arena—a modular and extensible benchmarking platform that, for the first time, systematically quantifies the impact of 33 distinct types of assumption violations on mainstream algorithms. Leveraging a large-scale simulation framework encompassing approximately 30 million experiments, the study comprehensively characterizes the robustness profiles of diverse algorithms and demonstrates that ensemble strategies substantially enhance generalization robustness. These findings provide both empirical grounding and practical guidance for achieving reliable causal discovery in real-world scenarios where assumptions may be imperfectly satisfied.

Abstract

Causal Discovery (CD) is a powerful framework for scientific inquiry. Yet, its practical adoption is hindered by a reliance on strong, often unverifiable assumptions and a lack of robust performance assessment. To address these limitations and advance empirical CD evaluation, we present TCD-Arena, a modularized, highly customizable, and extendable testing kit to assess the robustness of time series CD algorithms against stepwise more severe assumption violations. For demonstration, we conduct an extensive empirical study comprising around 30 million individual CD attempts and reveal nuanced robustness profiles for 33 distinct assumption violations. Further, we investigate CD ensembles and find that they have the potential to improve general robustness, which has implications for real-world applications. With this, we strive to ultimately facilitate the development of CD methods that are reliable for a diverse range of synthetic and potentially real-world data conditions.

Problem

• Causal Discovery
• Time Series
• Robustness
• Assumption Violations
• Empirical Evaluation

Innovation

• causal discovery
• time series
• robustness evaluation
• assumption violation
• ensemble methods