Generalized Linear Mixed Models, Fall 2015 (KU)

Upcoming Workshops:

Thank you for visiting my course notes. Here are some upcoming opportunities to learn from me and my colleagues in person:

Currently no workshops are planned

Course Information

Instructor:Dr. Jonathan Templin
Office Phone:785-864-5714
Office:614 Joseph R. Pearson Hall
Office Hours:Fridays: 10:30am-12:30pm or by appointment
Course Meeting Time:Wednesdays from 1:30pm-4:20pm
Course Meeting Location:143 Joseph R. Pearson Hall

Brief Course Description

In this course, contemporary methods for the analysis of data that are not normally distributed are presented. Generalized linear mixed models are linear models (regression/analysis of variance) that map independent or predictor variables onto the any type of outcome space of dependent variable(s). In general, such models involve the choice of distribution for the dependent variable(s) and a link function to map predictors onto key parameters of the dependent variable(s) distribution(s). Topics covered include logistic regression (or linear models for ordinal multinomial distributions), nominal distributions (or unordered multinomial distributions), count distributions (i.e., Poisson and negative binomial regression), and distributions for censored, skewed, or otherwise irregular continuous data. Random effects or latent variables (effects for nested data of all sorts) are included for each type of model.

The course will use the R statistical program with R Studio and SAS:

SAS is available in several ways:
1. For purchase online at the KU webstore (Windows OS only)
2. For free with the use of a virtual machine (all OS) by visiting
3. For free on the KU Advanced Computing Facility Supercomputers. See

For all other specific information regarding general course policies, course evaluation rubrics, and grading systems, please see the course syllabus at the link below.

Course Materials

Helpful R Links and Resources

Books: R for Data Science (Grolemund and Wickham, 2015)
R for SAS and SPSS Users by Muenchen (2007)
An introduction to R by Venables et al. (2013)
aRrgh: a newcomer's (angry) guide to R
DataCamp Introduction to R
Cookbook for R
Quick References:Reference Card #1 (various authors, noted on card)
Reference Card #2 (various authors, noted on card)

Helpful SAS Links and Resources

Book's SAS Files:SAS Example and Data Files by Stroup (2012)

Tentative Schedule of  Course Topics

DateTopic & Content LinkRequired ReadingAssignments
26-AugIntroduction to Generalized Linear Mixed Models and Mathematical StatisticsAppendices A and B
Homework #0 (Extra Credit)
Due 2 Sep at 1:30pm
2-SeptIntroduction to GLMMs/Models with MatricesChapter 1
#1: Matrices and Distributions (Due 23 Sept 2015)

Homework Data Set
9-Sept Research Design / Fundamentals of GLMMs
Chapter 2
16-SeptFundamentals of GLMMsChapter 3
23-SeptEstimation of GLMMs (Part 1)Chapter 4
30-SeptInformation, Standard Errors, and Newton-Raphson/Fisher ScoringNone#2: Maximum Likelihood (Due 21 Oct 2015)
7-OctThe Exponential FamilyNone
21-OctEstimation of Generalized Linear ModelsChapter 4#3: Estimation of Generalized Linear Models (Due 11 Nov 2015)
28-OctGaussian Linear Mixed ModelsChapter 4
4-NovML and REML EstimationChapter 4
11-NovEstimation of GLMMs; Inference in GLMMsChapter 5#4: REML Estimation (Due 2 Dec 2015)
18-NovExamples of Gaussian GLMMs
9-DecExamples of Non-Gaussian and Doubly Generalized GLMMs


Lee, Y., & Nelder, J. A. (1996). Hierarchical generalized linear models. Journal of the Royal Statistical Society. Series B (Methodological), 619-678. Paper link from journal.

Lee, Y., & Nelder, J. A. (2006). Double hierarchical generalized linear models (with discussion). Journal of the Royal Statistical Society: Series C (Applied Statistics), 55(2), 139-185. Paper link from journal.

Olsen, M. K., & Schafer, J. L. (2001). A two-part random-effects model for semicontinuous longitudinal data. Journal of the American Statistical Association, 96(454), 730-745. Paper link from journal.

Rose, C. E., Matin, S. W., Wannemuehler, K. A., & Plikaytis, B. D. (2006). On the use of zero-inflated and hurdle models for modeling vaccine adverse event count data. Journal of biopharmaceutical statistics, 16(4), 463-481. Paper link from journal.

Stroup, W. (2012). Generalized Linear Mixed Models: Modern Concepts, Methods and Applications. Chapman & Hall.