NATIONAL CHIAO TUNG UNIVERSITY

INSTITUTE OF STATISTICS

 

LINEAR MODELS

SPRING 2004

 

 

 


Instructor:

Guan-Hua Huang, Ph.D.

 

Office: 423 Assembly Building 1

 

Phone: 03-513-1334

 

Email: ghuang@stat.nctu.edu.tw

Class meetings:

Tuesday 9:00 am -12:00 pm at 406 Assembly Building 1

Office hours:

By appointment

Class website:

http://www.stat.nctu.edu.tw/faculty/ghuang/course/linearmodels04/

Credit:

Three (3) credits

 

COURSE SUMMARY

 

The objects of his course are

 

Ÿ             To present regression methodologies for continuous, categorical and count responses unified under the framework of generalized linear 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 continuous, binary, polytomous and count data; and estimating functions.

 

HANDOUTS AND TEXTBOOKS

 

Handouts corresponding to each lecture will be available on the class website before each class. There is one required textbook for this course and reading assignments will be made primary in this book:

 

McCullagh P. and Nelder J.A. (1989). Generalized Linear Models, 2nd edition. Chapman and Hall.

 

PREREQUISITES

 

Students are expected to have background on undergraduate probability, mathematical statistics, and linear regression.

 

METHOD OF STUDENT EVALUATION

 

The course grade will be based on five homework assignments (50%), one midterm exam (20%), and one final exam (30%).

 

COURSE OUTLINE

 

Readings refer to: McCullagh P. and Nelder J.A. (1989): Generalized Linear Models, 2nd edition.

 

Module

Topic

Reading

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 goodness-of-fit

Ÿ         Iterative reweighted least squares algorithm

Chapter 2

3

Analysis of continuous data

Ÿ         Error structure

Ÿ         Systematic component

Ÿ         Model formulae

Ÿ         Normal distribution – likelihood function

Ÿ         Estimation

Chapter 3

4

Analysis of binary data

Ÿ         Binomial distribution

Ÿ         Link functions

Ÿ         Case-control applications

Ÿ         Over-dispersion

Chapter 4

5

Analysis of polytomous data

Ÿ         Multinomial distribution

Ÿ         Model for nominal scales

Ÿ         Model for ordinal scales

Ÿ         Nested or hierarchical response scales

Chapter 5

6

Analysis of count data

Ÿ         Poisson distribution

Ÿ         Log-linear model for contingency tables

Ÿ         Connection with multinomial model

Ÿ         Poisson regression – application to cohort studies

Chapter 6

7

Quasi-likelihood function and estimating functions

Ÿ         Construction of quasi-likelihood function

Ÿ         Optimal estimating functions

Ÿ         Some applications of estimating functions

Chapter 9