STAT 6841, Categorical Data Analysis with Generalized Linear Models, Winter 2005

Monday
Wednesday

In Chapter 1, do exercise #1, 2, and 3. No need to submit.

Read sections 1.3-5.

Download (http://www.r-project.org/ ) and install R and/or install SAS. Brush up your computation skill

HW #1. In Agresti Chapter 1, do exercises #1, 2, 3, 4, 5 (Also, find the P-value and confidence intervals using Wald test as well; you need to read section 1.4.1 and 1.4.2)
Due Wednesday (1/12)

Reading Assignment : sections 1.4.3, 1.5.1-1.5.3, 2.1, 2.2
Until Wednesday (1/12), but preferrably until Monday (1/10)

Ch.1 Introduction: Distributions and inference for categorical data

Dr. Norton will teach on Monday (1/10) class

This week's material will be the most mathematical of what you'll learn from the course this quarter. From week 2, we will cover more application-oriented stuff using a computer package and real data.

HW #2. In Agresti Chapter 2, do exercises #1, 2, 3, 4, 8
Due Wednesday (1/19)

Reading Assignment : sections 2.4, 3.1-3.3, 3.5 (you can skip sections with *)

Also, install SAS and/or R by this week
Brush up your SAS data step/R

HW #3. In Agresti Chapter 3, do exercises #1, 4 (don't do question b), 18. You can do it by hands but I encourage you use SAS or other software to do that.
Due Wednesday (1/26)

Reading Assignment : sections 3.5, 3.6, 3.7 (you can skip sections with *)

Ch.3 Inference for contingency tables

Sample SAS code

Sample R code:
# heart attack / aspirin data
x<-as.table(matrix(c(189, 10845, 104, 10933), byrow=TRUE,nrow=2))
summary(x)
(x[1,1]*x[2,2])/(x[1,2]*x[2,1]) # odds ratio
# repeat the above for the lung cancer data
x<-as.table(matrix(c(189, 10845, 104, 10933), byrow=TRUE,nrow=2))

Monday
Wednesday

(3.1, 3.2, 3.3, 3.5, 4.1, 4.2, 4.3, These files are large!)

HW #4. In Agresti Chapter 3, do exercises #3.13 (reproduce the SAS result yourself) In Chapter 4, do 4.1, 4.2, 4.5, 4.7 (a)-(c), 4.8 (reproduce the SAS results yourself). Use SAS or R.
Due Wednesday (2/2)

Reading Assignment : sections 4.5, 5.1-5.3 (you can skip sections with *)

Also think about which section you want to do for projects after the midterm (see below).

HW solutions HW#1, HW#2

Ch.4 Logistic regression
SAS program for logistic/loglinear GLMs
R program for GLMs
crab.dat data file
snoring.dat data file

Project Assignment and Guidelines

Reading Assignment : sections 5.4-5.5

Midterm #1 (Feb7th) will be take-home. More details later.

6.1 strategies in model selections
Mi Lam, Dean Pangelinan (ppt, ppt, Word doc)

Reading Assignment : sections 6.1-4

Midterm Use crab2.dat and challenger.dat

In-Class sas codes

6.2 logistic regression diagnostics
Carol Ellis, Rommel Vives (pdf)

6.3 inference about conditional associations in 2x2xK tables
Gary Gongwer, Demeke Kasaw (ppt)

HW #5. In Agresti Chapter 6, do exercises #1-8 (Due Wednesday 2/23)

Reading Assignment : sections 7.1-3, 10.1-2

Midterm solutions removed for MS exam

* If your SAS system hangs while running ODS (Output Delivery Systems) graphics in solution.sas, fix the sas config file following instructions in this page

7.1 nominal responses: baseline-category logit models
Kathy Fung, Lin Zhang (ppt)

7.2 ordinal responses: cumulative logit models
Dan Sultana, Denise Hum (ppt, ppt)

HW #6. In Agresti, do exercises #7.1-7.3 and #10.1
(Due Wednesday 3/2)

Reading Assignment : sections 8.1-4, 9.1-4. (Optional: 11.1-2)

HW solutions HW#5

Ch. 7 Logit Models for Multinomial Responses

Ch. 10 Models for Matched Pairs

8.2 loglinear models for independence and interacton in 3-way tables
Robert Lagier, Veronica Estrada (ppt)

9.2 model selections and comparisons
Alvin Hsieh, Yumi Kubo (ppt, survey.sas, hr_survey.sas)

HW #7. In Agresti, do exercises #8.1, 8.5
(Due Wednesday 3/9)

Reading Assignment:

HW solutions HW#6

Ch. 8 Loglinear Models for Contingency Tables

Ch. 11 Analyzing Repeated Categorical Response Data

9.4 modeling ordinal associations
Bin Hu, Roanna Gee (rgproj.ppt, rgmovie.ppt, CDA table 9.3.xls, Presentation2.ppt)

Review Session

HW solution HW #7 (6841-hw7.sas)

Ch. 12 Random Effects: Generalized Mixed Models for Categorical Resonses

More topics if time permits