GENERALIZED
LINEAR (AND ADDITIVE) MODELS
FALL 2009
Instructor: 
GuanHua Huang, Ph.D. 

Office: 423 Joint Education Hall 

Phone: 035131334 

Email: ghuang@stat.nctu.edu.tw 
Class meetings: 
Thursday 9:0012:00 at 407 Joint Education Hall 
Office hours: 
By
appointment 
Class website: 

Credit: 
Three (3) credits 
The objects of his course are
Ÿ To present regression methodologies for categorical and count responses
unified under the framework of generalized linear models and generalized
additive models.
Ÿ To familiarize the usage of statistical software implementing these regression
methodologies.
Ÿ To provide references for your future research.
Topics include review of likelihood inference and
large sample test statistics, generalized linear models framework, analysis of
binary, polytomous and count data, smoothing, additive models and generalized
additive models.
Handouts corresponding to each lecture
will be available on the class website before each class. Reading assignments are from the following two books:
Ÿ
McCullagh P. and Nelder J.A. (1989). Generalized
Linear Models, 2^{nd} edition. Chapman and Hall.
Ÿ
Hastie T.J. and Tibshirani R.J. (1990). Generalized
Additive Models. Chapman and Hall.
Students
are expected to have background on undergraduate probability, mathematical
statistics, and linear regression.
The course grade will be based on four homework assignments (50%), one midterm
exam (20%),
and one final exam (30%).
COURSE OUTLINE
McCullagh
P. and Nelder J.A. (1989): Generalized Linear Models, 2^{nd} edition.
(McCullagh & Nelder)
Hastie T.J.
and Tibshirani R.J. (1990). Generalized Additive Models. (Hastie &
Tibshirani)
Module 
Topic 

1 
Review Ÿ
Likelihood
function and some basic properties Ÿ
Exponential
family Ÿ
Three large
sample test Ÿ
Conditional
likelihood inference 

2 
Generalized
linear models (GLM) Ÿ
The origins of GLM Ÿ
Systematic and random components of GLM Ÿ
Some statistical properties of GLM Ÿ
Maximum likelihood estimation versus weighted least squares Ÿ
Deviance  a measure of goodnessoffit Ÿ
Iterative reweighted least squares algorithm 
McCullagh
& Nelder Chapter 2 
3 
Analysis
of binary data Ÿ
Binomial distribution Ÿ
Link functions Ÿ
Casecontrol applications 
McCullagh
& Nelder Chapter 4 
4 
Analysis
of polytomous data Ÿ
Multinomial distribution Ÿ
Model for nominal scales Ÿ
Model for ordinal scales Ÿ
Nested or hierarchical response scales 
McCullagh
& Nelder Chapter 5 
5 
Analysis
of count data Ÿ
Poisson distribution Ÿ
Loglinear model for contingency tables Ÿ
Connection with multinomial model Ÿ
Poisson regression – application to cohort studies 
McCullagh
& Nelder Chapter 6 
6 
Smoothing Ÿ
Bin smoothers Ÿ
Kernel smoothers Ÿ
Splines 
Hastie
& Tibshirani Chapter 2 
7 
Additive
models Ÿ
Fitting additive models Ÿ
Estimating equations for additive models Ÿ
Solutions to the estimating equations 
Hastie
& Tibshirani Chapters 4, 5 
8 
Generalized
additive models (GAM) Ÿ
Local scoring for GAM Ÿ
Semiparametric GAM Ÿ
Inferences Ÿ
Smoothing parameter selection 
Hastie
& Tibshirani Chapter 6 