Robust Causal Discovery in Real-World Time Series with Power-Laws

2025/07/16 by Matteo Tusoni, Giuseppe Masi, Tusoni, Matteo +9
Computer Science · Engineering · #Bayesian Modeling and Causal Inference #Data Analysis #FOS: Computer and information sciences #FOS: Physical sciences #Fault Detection and Control Systems #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Other Statistics (stat.OT) #Statistics and Probability (physics.data-an) #Time Series Analysis and Forecasting

paper · pdf · doi:10.48550/arxiv.2507.12257

openalex publication_date 2025/07/16 · openalex created_date 2025/10/09 · openalex updated_date 2026/07/28

Abstract

Exploring causal relationships in stochastic time series is a challenging yet crucial task with a vast range of applications, including finance, economics, neuroscience, and climate science. Many algorithms for Causal Discovery (CD) have been proposed; however, they often exhibit a high sensitivity to noise, resulting in spurious causal inferences in real data. In this paper, we observe that the frequency spectra of many real-world time series follow a power-law distribution, notably due to an inherent self-organizing behavior. Leveraging this insight, we build a robust CD method based on the extraction of power-law spectral features that amplify genuine causal signals. Our method consistently outperforms state-of-the-art alternatives on both synthetic benchmarks and real-world datasets with known causal structures, demonstrating its robustness and practical relevance.

Citations

• Are you doing better than random guessing? A call for using negative controls when evaluating causal discovery algorithms
• Robust Time Series Causal Discovery for Agent-Based Model Validation
• Causal Discovery in Semi-Stationary Time Series
• CausalTime: Realistically Generated Time-series for Benchmarking of Causal Discovery
• Causal Discovery from Temporal Data: An Overview and New Perspectives
• Rhino: Deep Causal Temporal Relationship Learning With History-dependent Noise
• DYNOTEARS: Structure Learning from Time-Series Data
• CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training
• Fast Causal Inference with Non-Random Missingness by Test-Wise Deletion
• The Do-Calculus Revisited
• Measurement of Linear Dependence and Feedback between Multiple Time Series

Related

• Causal Discovery on Irregular Time Series
• Causal discovery for time series with latent confounders
• Efficient Causal Discovery for Autoregressive Time Series
• Causal Discovery in Multivariate Time Series through Mutual Information Featurization
• Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data