QMEG Masterclass: Introduction to Causal Inference with Machine Learning
Learn Double Machine Learning: use ML to estimate valid causal effects, with a hands-on Python example. No prior causal knowledge needed.
Many research questions are causal, yet standard machine learning models are built for prediction, not for estimating the effect of an intervention. Double Machine Learning (DML) bridges this gap, letting researchers use flexible ML models to control for complex confounding while still producing valid effect estimates and confidence intervals.
This online seminar introduces the core ideas behind DML, including orthogonalisation and cross-fitting, with a worked example in Python. We will also cover the key assumptions and limitations.
Suitable for researchers and postgraduate students with basic regression knowledge; no prior causal inference experience needed.
Good to know
Highlights
- 2 hours
- Online