Time-Series Causal Discovery

Ask about this topic

Overview

Time-series causal discovery aims to distinguish directional influences between variables, including effects at different time lags and relationships that change over time. Approaches include Granger-causality tests and vector autoregressions, conditional-independence methods such as PCMCI, score-based and constraint-based graph search, and neural models that learn sparse or changing graphs. Researchers apply these methods to settings such as climate records, clinical measurements, industrial telemetry, and irregular event streams, while developing tools for latent confounding, non-stationarity, and limited observations. A central challenge is that predictive dependence does not by itself establish causation: conclusions rely on assumptions that may fail, and benchmarks must test recovery under realistic violations and interventions.

Latest papers

CardsList

65 papers9 papers in the last four weeks, up 50% on the four weeks before. 0.1% of all new papers.

  1. Uncertainty Quantification in Federated Granger Causality Learning

    Feb 13, 2026Ayush Mohanty, Nazal Mohamed, Nagi GebraeelTime-Series Causal DiscoveryCausal Discovery

    Georgia Institute of Technology Atlanta, GA 30332

  2. DCD: Decomposition-based Causal Discovery from Autocorrelated and Non-Stationary Temporal Data

    Feb 1, 2026Muhammad Hasan Ferdous, Ian Stockwell, Md Osman GaniTime-Series Causal DiscoveryCausal Discovery

    Causal AI Lab, Department of Information Systems, University of Maryland, Baltimore County, Baltimore, MD, USA

  3. Learning Granger Causality under Latent Confounding via Intervention-Induced Heterogeneity

    Oct 22, 2025Ziyi Zhang, Shaogang Ren, Xiaoning Qian +1Time-Series Causal DiscoveryCausal Discovery

    Texas A&M University · University of Tennessee at Chattanooga · Brookhaven National Laboratory

  4. Spectral structure learning for clinical time series

    Feb 17, 2025Ivan Lerner, Anita Burgun, Francis BachMulti-Output Gaussian ProcessesTime-Series Causal Discovery

    Université Paris Cité, Inria, Inserm, HeKA, F-75015 Paris, France · Service d’informatique biomédicale, AP-HP, Hôpital Georges Pompidou, F-75015 Paris, France · SIERRA, Inria Paris, F-75015 Paris, France

  5. Granger Causality in Extremes

    Jul 12, 2024Juraj Bodik, Olivier C. PascheTime-Series Causal DiscoveryCausal Inference

    Faculty of Business and Economics, University of Lausanne, Switzerland · Department of Statistics, UC Berkeley, California, USA · Research Institute for Statistics and Information Science, University of Geneva, Switzerland +1