Robust causal inference using non-randomized longitudinal data

Abstract

While RCTs are, and will likely remain, the gold standard for causal inference, there are times when RCTs are not possible, but where causal inference is desirable. This seminar provides an overview of robust methods for estimating causal effects using observational data, and particular the method targeted maximum likelihood estimation.

Philip J Clare

Biostatistician at the Prevention Research Collaboration, University of Sydney.