STA 312F 2010

Topics in Statistics: Categorical Data Analysis

University of Toronto at Mississauga, Fall 2010

Lecture: Tuesday 10:10-12:00 in NE 172 and Thursday 11:10-12:00 in SE 1104.

Tutorial: Friday 11:10-12:00 in CC 2212

(A quiz will be given in each tutorial.)

This page is under construction, and will be updated frequently throughout the course.

I forgot to post the review slide show again. Here it is. I'm sorry about that.

Office hours for the final exam are:

• Tuesday Nov 30: 10am - 12 noon (Jerry)
• Thursday Dec. 2nd: 10am - 12 noon (Jerry)
• Tuesday Dec. 7th: 10am - 12 noon (Christine)
• Wednesday Dec 8th: 10 am - 1pm (Jerry)

Information about the final exam, including the final assignment, is posted below. It's now the second item in the list.

• Course Outline
• About the Final Exam: This includes the final computer assignment and a link to the formula sheet that will be used on the exam. If any additions are made to this document, you will see a notification near the top of this page.
• Christine's Website for this course.
• Computing Information and Links
• Homework Assignments
  1. Assignment 1: Quiz on Friday Sept. 17 in tutorial.
  2. Assignment 2: Quiz on Friday Sept. 24 in tutorial.
  3. Assignment 3: Quiz on Friday Oct 1st in tutorial.
  4. Assignment 4: Quiz on Friday Oct 8th in tutorial.
  5. Assignment 5: Quiz on Friday Oct 15th in tutorial.
  6. Assignment 6: Quiz on Friday Oct 22nd in tutorial.
  7. Assignment 7: Quiz on Friday Oct 29th in tutorial.
  8. Assignment 8: Quiz on Friday Nov. 5th in tutorial.
  9. Assignment 9: Quiz on Friday Nov. 19th in tutorial.
  10. Assignment 10: Quiz on Friday Nov. 26th in tutorial.
• Handouts and Overheads: You may want to print these and bring to class so you can write notes on them. Some of the overheads are in Powerpoint format. If you don't have Microsoft Office, you can use Open Office, which is a free download from www.openoffice.org.
  1. Introduction: Powerpoint slides of opening lecture. Please read Chapter One of the text for another overview.
  2. Multinomial Distributon Lecture(s): This important background material is not in the text.
  3. Chapter Two
     • Log-linear Part One
     • Log-linear Part Two
  4. Chapter Three: Log-linear Part Three If this is giving you trouble, there is also a version in an older Powerpoint format, with a larger file size. Click here for the version in the older format.
  5. Chapter Four: Log-linear Part 3.5 (pdf format)
  6. A short addition: Marginal tables with R.
  7. Berkeley Graduate Admissions data. This is still part of Chapter 4.
  8. Chapter Five
     • Log-linear Part Four (short)
     • Analysis of the Detergent Data given in Table 5-1.
     • Corrected Analysis of the Detergent Data (Remember, I accidentally left out a main effect specification.)
     • Brief exploration of what R does when you leave out a main effect specification.
  9. Chapter Six
     • Log-linear Part Five: Fixed Margins and Logit Models
     • Ordinary multiple regression with dummy variables (review?)
     • Logistic Regression Part 1
     • Low Birth Rate study: Logistic regression example.
     • Low Birth Rate Part 2: Doing logistic regression tests with log-linear models.
     • Stepwise Part 1: Logistic regression and log-linear models.
     • Poisson Regression
     • Poisson Regression on the language data
     • Logistic Regression Part 2: More than 2 outcomes.
  10. SAS
      • Textbook Chapter on unix and SAS
      • SAS Logistic Regression Manual (big)
      • Birth Weight Part 1: Describing the data with SAS. You might want to just print pages 1-7 at most, or page 1 at least.
      • Birth Weight Part 2: Logistic Regression with SAS
      • Wald Tests: This is a mildly theoretical slide show. It is helpful at this point because proc logistic places such a heavy emphasis on Wald tests. Score tests are also explained. The first three slides are repeated from earlier presentations, and you might decide not to print them.
      • Birth Weight Part 3 with SAS: More logistic regression
      • Birth Weight Part 4 with SAS: Miscellaneous
      • Logistic regression with more than 2 outcomes: Math data using SAS
      • Logistic regression with an ordered categorical response variable: Math data using SAS
• Data sets from lecture and homework.