Dear Applied Statistics Community, Please join us this Wednesday, when Thomas Richardson--Department of Statistics, University of Washington--will present "Analysis of the Binary Instrumental Variable Model", work that is joint with Jamie Robins, Harvard School of Public Health. Thomas provided the following abstract: In this talk I consider an instrumental variable potential outcomes model in which the instrument (Z), treatment (X) and response (Y) are all binary. It is well known that this model is not identified by the observed joint distribution p(x,y,z). Consequently many statistical analyses impose additional untestable assumptions or change the causal estimand of interest. Here we take a different approach, directly characterizing and graphically displaying the set of distributions over potential outcomes that correspond to a given population distribution p(x,y,z). This provides insights into the variation dependence between the partially identified average causal effects for various compliance groups. The analysis also leads directly to re-parametrization that may be used for Bayesian inference and the development of models that incorporate baseline covariates. The Applied Statistics Workshop meets each Wednesday at 12 noon in K-354 CGIS-Knafel (1737 Cambridge St). The workshop begins with a light lunch and presentations usually start around 1215 and last until about 130 pm. Cheers Justin Grimmer