Machine Learning and Causal Effects
10 - 12 JUNE 2024

Eduardo Fé, Eleonora Iob, and Andrea Aparicio Castro

Course description

This intermediate course offers a comprehensive understanding of the links between machine learning and traditional methods in causal inference. This course was designed for participants that are familiar with causal inference and the fundamentals of Machine Learning.

Key topics covered:

  • Potential Outcomes: Understand the concept of potential outcomes and how machine learning can be applied to estimate causal effects in observational studies.
  • Matching and Instrumental Variables: Explore advanced methods such as matching and instrumental variables to address confounding variables and strengthen causal inference models.
  • Trees and Regularisation: Learn how decision trees and regularisation techniques can be integrated into causal analysis, providing participants with tools to handle complex datasets and improve model performance.

There are two main learning objectives:

  • To develop an understanding of how Machine Learning is being applied for causal inference in real-world studies.
  • To be able to implement some of the available methods in R.

There are three scheduled live online sessions, each consisting of a lecture and a hands-on computer practical session. During the practicals, participants will be provided with materials to implement the methods discussed in the respective lecture. Solutions will also be given to enable them to easily apply what they learn to their own work.




Monday 10th June 2024


Live session and practical 1

Tuesday 11th June 2024


Live session and practical 2

Wednesday 12th June 2024


Live session and practical 3

Participants should be familiar with causal inference, including key concepts such as confounding, treatment effects, and model evaluation is essential. In addition, attendees should have basic knowledge of Machine Learning to the level of the following suggested reading list:

  • James, G., D. Witten, T. Hastie, R. Tibshirani (2021) Introduction to Statistical Learning with Applications in R, 2nd Edition. Springer Texts in Statistics. Available for download at: https://www.statlearning.com/

£90 per person and £45 for PhD students

This is a UKRI funded project offering rigorous training in longitudinal data science. Please note that this training is NOT available to undergraduate or Masters students.