Stats 201C    Advanced Modeling and Inference

Course description: Introduction to advanced topics in statistical modeling and inference, covering two groups of topics:
(A) Statistical inference for incomplete data and hidden variable models;
(B) Sparse regularization for linear, generalized linear and graphical models.

Main topics:
(1) Incomplete data and the EM algorithm: EM and its properties, incomplete multivariate normal data.
(2) Hidden variable models: Mixture modeling, EM clustering, stochastic block models, variational EM.
(3) Sparse linear models: Lasso, group lasso, high-dimensional inference, de-biased lasso, estimator augmentation.
(4) Sparse graphical modeling: Gaussian graphic models, covariance selection, structure learning of directed acyclic graphs, causal inference with intervention data.


Link to the course CCLE site for lecture notes, homework assignments, etc.