The goal of this course is introduce various statistical learning methods that are widely used in medical and public health research. The topics in this course include penalized linear regression, dimension reduction regression, logistic regression, discriminant analysis, support vector machines, tree-based methods, principal component analysis and clustering. The resampling procedures including cross-validation, bootstrapping and permutation tests, which are important in model evaluation and inference, will be introduced as well.

Integration of linear models with experimental design and sampling, considering applications with unbalanced and missing data. In-depth discussion of mixed models including parameterizations, analysis of covariance, unequal numbers of observations per cell, missing cells. Repeated measure designs and longitudinal data analysis emphasized, with many examples illustrated using SAS. Prerequisite: BIOST&EP 640 and 691F or equivalent. Recommended: Bioepi 741 and 744.

This is not available at this time

Principles of statistics applied to analysis of biological and health data, evaluation of public health and clinical programs.

Description not available at this time