EuroCIM workshop: Causal machine learning
Evaluating treatment effects using observational data increasingly requires adjustment for a high-dimensional set of covariates in order to control confounding. This is the result of a lack of comparability between treated and untreated subjects in possibly many (pre-treatment) factors that are also related to outcome. While such adjustment is routinely achieved via parametric modelling, it is not entirely satisfactory as model misspecification is likely, and even relatively minor misspecifications over the observed data range may induce large bias in the treatment effect estimate. Unsurprisingly, machine learning methods are increasingly being used to assist in this task.
This workshop will cover the use of machine learning in causal effects estimation. It should also be of interest to researchers who rely on variable selection procedures in their data analyses.
This workshop will consist of 2 parts. Participants can register for either parts or both.
Part 1: Introductory workshop (09.00 - 11.00 BST)
- 09.00 - 09.45: Session 1 - Introduction to causal machine learning
- 09.45 - 10.00: Break
- 10.00 - 10.45: Session 2 - Causal machine learning in action
- 10.45 - 11.00: Q&A
Please note that this session is now fully booked.
Part 2: Advanced workshop (13.00 - 16.00 BST)
- 13.00 - 13.45: Session 3 - Calculating the canonical gradient of an estimand
- 13.45 - 14.00: Break
- 14.00 - 14.45: Session 4 - Hands-on calculating the canonical gradient of an estimand
- 14.45 - 15.00: Break
- 15.00 - 15.45: Session 5 - The von mises expansion
- 15.45 - 16.00: Q&A
EuroCIM Conference
This workshop will be followed by a conference on “methods” on 19 May.
- Please note that the sessions will not be recorded.
- Registration disclaimer: By entering your contact details, you are agreeing to receive follow up communication specific to this workshop from the LSHTM Centre of Statistical Methodology. Communications will include teaching materials sent in advance and following the workshop.
- For more information on how we use your data, please see our and our .
Admission
Contact
