Longitudinal Data Analysis
Course covers statistical issues surrounding estimation of effects using data on units followed through time. Course emphasizes a regression model approach for estimating associations of disease incidence modeling, continuous outcome data/linear models & longitudinal extensions to nonlinear models forms (e.g., logistic). Course emphasizes complexities that repeated measures has on the estimation process & opportunities it provides if data is modeled appropriately. Most time is spent on 2 approaches: mixed models based upon explicit (latent variable) maximum likelihood estimation of the sources of the dependence, versus empirical estimating equation approaches (generalized estimating equations). Primary focus is from the analysis side.
Topics
- After successfully completing the course, you will be able to: • frame data science questions relevant to longitudinal studies as the estimation of statistical parameters generated from regression, • derive consistent statistical inference in the presence of correlated, repeated measures data using likelihood-based mixed models and estimating equation approaches (generalized estimating equations; GEE), • interpret the regression output, including both coefficients and variance components and • implement the relevant methods using R.
Official catalogue course objectives and student learning outcomes
Sources
- Course catalogue ↗2026-2027