Causality

Theme Overview

Scientific inquiry often revolves around uncovering true causal relationships amidst a sea of correlations. While randomized controlled trials serve as the gold standard, practical constraints often limit their feasibility. The growing field of causal inference aims to develop a rigorous framework for establishing causal effects from observational data and its combination with limited controlled experimentation. 

This enterprise involves carefully formulating the causal question of interest, explicitly stating the assumptions under which the answers can be interpreted causally, and providing algorithms and inferential tools for deducing the consequences of such assumptions. With the growing demand for answering "what if" questions about how the world should behave when intervened upon, research in causal inference often involves developing novel statistical and computational tools.The field also encompasses methodology for uncovering the underlying mechanisms driving phenomena, collectively referred to as causal discovery. 

The Causality group at UCL Statistical Science works on causal discovery, Bayesian causal inference, causal machine learning and counterfactual prediction and fairness amongst other topics. Furthermore, the department actively engages in interdisciplinary collaborations, applying causal inference methodologies to domains such as Health, Social policy, economics, and beyond. Through these collaborations, researchers at UCL Statistical Science Department contribute to evidence-based decision-making and policy formulation, tackling real-world challenges with rigorous statistical approaches.

In addition to research, the department is committed to education and training, offering cutting-edge courses and workshops on causal inference for both students and professionals. By disseminating knowledge and fostering expertise in causality, UCL's Statistical Science Department continues to be at the forefront of advancing our understanding of cause-and-effect relationships in the modern world.

Theme Members

• Gianluca Baio
• Karla Diaz-Ordaz (Theme Lead)
• Elinor Jones
• Brieuc Lehmann
• Ioanna Manolopoulou
• Giampiero Marra
• Kayvan Sadeghi
• Ricardo Silva
• Terry Soo

Recent and Upcoming Events

• Frontiers in Statistical Science - IMSS Annual Lecture: Prof Andrea Rotnitzky (University of Washington), UCL, June 2024
• The UCL Centre for Data Science Causal Inference Symposium, London Mathematical Society, November 2023

Current and Recent Externally Funded Projects

• CHAI - Causality in Healthcare AI with Real Data, EPSRC AI Hub EP/Y028856/1, February 2024 - January 2029, Co-Is: Silva and Diaz-Ordaz
• Fairness in Clinical Prediction Models, The Alan Turing Institute, September 2023 - March 2025, PI: Lehmann, Co-Is: Silva and Diaz-Ordaz
• The Causal Continuum - Transforming Modelling and Computation in Causal Inference, EPSRC Open Fellowship EP/W024330/1, October 2022 - September 2027, PI: Silva
• Robust Factorial Causal Predictions with Observational and Interventional Data, ONR N62909-19-1-2096, Office for Naval Research, January 2021 - October 2023, PI: Silva