Estimating the effect of a treatment on some outcome of interest is an important goal in many scientific fields. Although randomized studies are the golden standard for estimating causal effects observational data play an important role in many scientific fields. Linkage of administrative register data based on the Swedish population forms a valuable data source for empirical scientists in both social science and medicine. For the analysis of observational data we can use statistical methods developed in the research field of Causal Inference. Here, propensity scores and instrumental variable methods are commonly used for the estimation of average causal effects.
Three renowned speakers will provide a comprehensive introduction to the subject together with highlights of recent advances in the research area. Applications where register data has been analyzed will also be demonstrated.
Date: March 8-12, 2015
Place: Borgafjäll, Sweden
Organizers: Umeå University, Swedish Statistical Society and Stat4Reg Lab.