Graduate | Spring 2018
The goals of this class are to train students to a) use Bayesian inference to build models that represent their understanding of the social phenomena of interest as well as competing or complementary representations, b) understand how to propagate information from the data, through the model, into a new computed representation of the process, c) interrogate a set of models to illuminate flaws that can be improved upon iteratively, d) use predictions to validate the usefulness of sets of models, and e) clearly understand the limits of Bayesian inference. To accomplish these goals, the class will introduce the Stan modeling language as well as guide students in writing some samplers from scratch.