Course Description
Multilevel and hierarchical models for social and biological sciences. Empirical Bayes, James-Stein, maximum likelihood, and Bayesian estimation of model parameters. Interpreting and diagnosing multilevel models, model building, and uncertainty assessment.
Athena Title
MULTILEVEL MODELS
Prerequisite
BIOS 7020 or STAT 6220 or STAT 6320
Semester Course Offered
Offered fall
Grading System
A - F (Traditional)
Course Objectives
A student completing this course should be able to: 1. Describe multilevel models as a generalization of linear regression models. 2. Describe multilevel models as a special case of a hierarchical Bayes model. 3. Build, fit, and evaluate multilevel and hierarchical models using classical and Bayesian methods. 4. Apply multilevel models to clustered sampling schemes, growth curves, random coefficient settings, and grouped experimental trials.
Topical Outline
1. Empirical Bayes and James-Stein Estimator 2. Multilevel Models 3. Model-checking using AIC/BIC/DIC 4. Maximum likelihood estimators, Fisher information, standard errors and confidence intervals 5. Bayesian statistics, the slogan, computation with conjugate priors 6. Random simulation and MCMC as a way to estimate a model, confidence intervals 7. Fitting and interpreting models using WinBUGS and rube() 8. Using WinBUGS for simulation tests