| Course ID: | STAT 8630. 3 hours. |
| Course Title: | Mixed-Effect Models and Longitudinal Data Analysis |
Course
Description: | Extensions of classical and generalized linear models with
emphasis on longitudinal data analysis. Course will focus on
linear mixed models, and marginal and mixed-effect versions of
generalized linear models for longitudinal discrete data.
Emphasis will be placed on the application of these models to
analyze real data. |
| Oasis Title: | LONG DATA ANALYSIS |
| Prerequisite: | STAT 8260 |
| Pre or Corequisite: | STAT 8620 |
Semester Course
Offered: | Offered spring semester every odd-numbered year. |
| Grading System: | A-F (Traditional) |
|
| Course Objectives: | Students will learn statistical models and methodology for the
analysis of continuous and discrete data for which the
assumptions of generalized linear models (GLMs) do not hold. In
particular, students will learn to analyze longitudinal data and
other data types for which the assumption of independence among
units fails. Students will learn the theory and application of
linear mixed-effect models and generalized linear mixed-effect
models, as well as other extensions of GLMs. Students will learn
the computational and algorithmic methods for fitting these
models as well as the implementation of these methods via
statistical software. |
| Topical Outline: | Introduction to longitudinal data; classical methods for
continuous longitudinal data; linear mixed models including
special cases, fitting and inference methodology, and
application to real data; generalized linear mixed models
including estimation, inference and application; marginal models
generalized linear models for longitudinal data and generalized
estimating equations; analysis in the presence of missing data.
Additional topics at the discretion of the instructor. |