STA 216 Generalized Linear Models Instructor: David Dunson 211 Old Chem, (NIEHS)
STA 216 Syllabus Topics to be covered: Definition of GLM: Components, assumptions and motivating examples The Basics: Exponential family, model fitting, and analysis of deviance Binary Data (Models): Link functions, parameter interpretation, & prior specification Binary Data (Computation): Approximations and MCMC algorithms
Topics (Page 2) Binary Data (Probit Models): Underlying normal structure and Albert & Chib Gibbs sampler Ordered Categorical Data: Probit models, common link functions, and examples Unordered Categorical Data: Multinomial choice models, common link functions and examples Log-Linear Models: Poisson distribution, parameter interpretation, over-dispersion and examples
Topics (Page 3) Discrete-Time Survival Models: Relationship with binary data models, convenient forms & examples Continuous-Time Survival: Proportional hazards model, counting processes & implementation Accounting for Dependency: Mixed models for longitudinal and multilevel data Multivariate GLMs: Generalized linear mixed models for multivariate response data
Topics (Page 4) Models for Mixed Discrete & Continuous Outcomes: Underlying normal & GLMM approaches Advanced Topics: Incorporating parameter constraints Hidden Markov and multi-state modeling Case Studies: Fertility and tumorigenicity applications Non- and semi-parametric methods Identifiability & improved methods for computation
Student Responsibilities: Assignments: Outside reading and problems sets will typically be assigned after each class (10%) Mid-term Examination: An in-class closed-book mid term examination will be given (30%) Project: Students will be expected to write-up and present results from a data analysis project (30%) Final Examination: The final examination will have both in-class (15%) & out of class problems (15%)