Symposium on Machine Learning for Causal Inference in the Health and Social Sciences

Date & Time

15 December 2023




London Mathematical Society, De Morgan House, London

Researchers from the University of York, London School of Hygiene and Tropical Medicine, University College London, and the Centre for Microdata Methods and Practice (Cemmap) are delighted to co-host a symposium on machine learning for causal inference in the health and social sciences.

The aim of the event is to facilitate a productive discussion among economists, statisticians, and health data scientists actively involved in developing and applying machine learning methods for causal inference. Our objective is to curate a programme that showcases methodological advancements and real-world case studies in economics, health care, and beyond. The preliminary programme includes talks on the estimation of heterogeneous treatment effects and optimal policy learning. The event will consist of three contributed sessions, one plenary session, and a concluding panel discussion (see list of speakers below, or download the full programme). The keynote speaker will be Professor Whitney Newey (Massachusetts Institute of Technology). Coffee and lunch breaks will provide opportunities for networking and discussing potential collaborations. 

Financial and other support for this event comes from the Medical Research Council, Wellcome Trust, Centre for Data and Statistical Sciences for Health (DASH) at the London School of Hygiene and Tropical Medicine, UCL Department of Economics, UCL Department of Statistical Science, Centre for Microdata Methods and Practice (Cemmap) and the Institute for Fiscal Studies


Keynote talk: 

  • Whitney Newey (Massachusetts Institute of Technology)

Session 1: Statistics:

  • Stijn Vansteelandt (University of Ghent) – IV-learner: learning conditional average treatment effects using instrumental variables
  • Karla Diaz Ordaz (UCL Department of Statistical Science) – Non-parametric variable importance measures for heterogeneous causal effects
  • Oliver Hines (QuantCo) – Causal inference with continuous treatments, a tale of two estimands

Session 2: Economics 

  • Martin Weidner (University of Oxford, Department of Economics) – A Neyman Orthogonalization Approach to the Incidental Parameter Problem
  • Julia Hatamyar (University of York, Centre for Health Economics) – Machine Learning for Difference-in-differences with Staggered Adoption and Dynamic Treatment Effect Heterogeneity
  • Liyang Sun (UCL, Department of Economics) – Empirical Welfare Maximization with Constraints

Session 3: Health Economics & health data science

  • Noemi Kreif (University of York, Centre for Health Economics) – Policy Learning with Rare Outcomes
  • Stephen O’Neill (London School of Hygiene and Tropical Medicine) – An approach for combining clinical judgment with machine learning to inform medical decision-making
  • David Glynn (University of York, Centre for Health Economics) -Integrating decision modelling and machine learning

Panel discussion: Making (causal) machine learning useful for policy: model validation and other challenges

  • Ioanna Manolopulou (UCL, Department of Statistical Science)
  • Richard Grieve (London School of Hygiene and Tropical Medicine)
  • Chris Harbron (Roche)
  • Stephen Hansen (UCL, Department of Economics)

Registrations are now closed for this event.