A Presentation by Fabrizia Mealli, PhD
Monday, May 22, 2017
Propensity score methods have proven to be exceptionally useful and straightforward tools for identifying a treatment effect when assignment to treatment is not random. Proper use of propensity scores in complex samples — clustered data, weighted survey data, multi-stage samples — requires additional considerations. Come learn more about emerging methodological work that responds to questions about the proper use of propensity score methods in complex samples!
Fabrizia Mealli is a Professor of Statistics at the Department of Statistics, Computer Science, Applications, University of Florence. She has held visiting positions at UCLA, Harvard University, LISER-Luxembourg, and ISER-University of Essex. Her research interests include causal inference, program evaluation, estimation techniques, missing data, and Bayesian inference. She has published in statistics, applied statistics, biostatistics, econometrics, economics, and demography journals. She is currently the Associate Editor for the Journal of the American Statistical Association T&M, Biometrics, the Annals of Applied Statistics.