EPI 531 STATISTICAL METHODS FOR ANALYSIS WITH MISSING DATA

Covers statistical methods for the analysis of missing data, including likelihood-based, weighted GEE, multiple imputation, and Bayesian approaches. Uses computational tools such as EM algorithm and Gibbs' sampler. Covers both ignorable and non-ignorable missing-data mechanisms as well as cross-sectional and longitudinal study designs. Primarily uses data arising from epidemiologic studies. Offered: jointly with BIOST 531.