Department of Biostatistics Seminar/Workshop Series

New developments in the estimation of treatment effects in observational studies: The use of data resampling with propensity score matching.

Laurie Samuels

Ph.D. Candidate in the department of Biostatistics

I will begin the seminar with a brief introduction to causal inference in observational studies, particularly in studies where there is incomplete covariate overlap between the treatment and control groups. I will then give an overview of the methods and main results from a recent article by Peter Austin and Dylan Small, "The use of bootstrapping when using propensity-score matching without replacement: a simulation study." (Statist. Med. (2014); http ://onlinelibrary.wiley.com/doi/10.1002/sim.6276/full). I will conclude with a discussion of possible extensions to Austin and Small's research.