README.md

Session 4: loglinear regression part 1

Learning objectives

1. Define log-linear models in GLM framework
2. Identify situations that motivate use of log-linear models
3. Define the Poisson distribution and the log-linear Poisson GLM
4. Identify applications and properties of the Poisson distribution
5. Define multi-collinearity

Outline

1. Brief review of GLMs
2. Motivating example for log-linear models
3. Poisson log-linear GLM
4. Note on multicollinearity

Reading: Vittinghoff textbook chapter 8.1-8.3

Lab

Learning Objectives

1. Simulate Poisson-distributed data with a relevant covariate
2. Fit a Poisson log-linear GLM
3. Create and interpret diagnostic plots for a log-linear GLM
4. Use analysis of deviance to compare two log-linear GLMs
5. Practice recoding and creating tables and plots