Causal inference for time series data

Many important research questions involve causation in systems for which direct experimentation is expensive, unethical or quite simply impossible. Examples include the Earth system, the human brain, socio-economic systems, epidemiology and industrial processes. Research on causal inference aims to provide both theoretical foundations and practical methods that can use domain knowledge and observational or experimental data to learn and quantify possible causal relationships between the variables of interest. Most real world data comes in the form of time series, which pose special difficulties for causal inference, and have been the subject of statistical study since the beginning of the 20th century. To deal with the challenges posed by time series data, several theoretical frameworks and practical methods have been developed, each based on their own set of underlying assumptions. Some of these approaches rely on the theory of continuous-time dynamical systems, others make use of results on stochastic processes or adapt the constraint-based approach to causality to time series data. Although recent works have made advances on several fronts, many challenges remain and causal inference for time series is arguably still a relatively underexplored area of research. 

In this workshop, we aim to bring together leading researchers and new investigators on causal inference for time series, as well as experts in dynamical systems and stochastic processes. 

• Submission Start: May 01 2023 11:59AM UTC-0 (April 30 2023 23:59 UTC-12)

• Submission Deadline: June 01 2023 11:59AM UTC-0 (May 31 2023 23:59 UTC-12) June 03 2023 11:59AM UTC-0 (June 2 2023 23:59 UTC-12)

• Author notification: July 5 2023 11:59AM UTC-0 (July 4 2023 23:59 UTC-12)

• Date of workshop: August 4 2023

cits.uai.2023@gmail.com