Estimating non-parametric propensity score and inverse probability of treatment weights when drawing causal inferences with observational data
The estimation of causal effects is a primary activity of many research studies. Controlled experiments are the gold standard for estimating such effects. However, experiments are often infeasible and only observational data, in which treatment assignment is not controlled by the researchers, are available for analysis. This workshop will provide an introduction to causal modeling with observational data using the potential outcomes framework and inverse probability of treatment (IPT) weights for the estimation of causal effects. It will also present step-by-step guidelines on how to estimate and perform diagnostic checks of IPT weights for testing the relative effectiveness of two or more interventions and the cumulative effects of time-varying interventions. Attendees will gain hands-on experience estimating IPT weights using generalized boosted models in R, SAS and Stata, evaluating the quality of those weights, and utilizing them to estimate intervention effects.
