
Course Outline
MATH 3330 A Fall 2001
Taught by: Professor Gene Denzel
Office: N615 Ross
Office Hours: M,W 910 and M 3:304:30, or by appointment
Email: Gene.Denzel@mathstat.yorku.ca
Web page: http://www.yorku.ca/lezned
Prerequisites: One of AS/SC/AK/MATH 2131 3.0, AS/SC/AK/MATH 2570 3.0, AS/SC/PSYC 2020 6.0, or equivalent; some acquaintance with matrix algebra (such as is provided in AS/SC/AK/MATH 1021 3.0, AS/SC/MATH 1025 3.0, AS/SC/MATH 1505 6.0, AS/AK/MATH 1550 6.0, AS/SC/MATH 2021 3.0, or AS/SC/AK/MATH 2221 3.0). Exclusions: AS/SC/MATH 3033 3.0, AS/SC/GEOG 3421 3.0, AS/SC/PSYC 3030 6.0, AS/ECON 4210 3.0, AK/PSYC 3110 3.0.Text: Applied Regression Analysis and Other Multivariable Methods, by Kleinbaum, Kupper, Muller, and Nizam (3rd Edition), Duxbury 1998.
Course content:
The course is intended as a thorough introduction to the use of linear models in statistical analysis, for students who have had at least two terms of statistics. We will be focussing on situations where we have one dependent variable and one or more explanatory variables. The material covered will be drawn from the textbook (see more details below), supplemented by some material from other sources. The emphasis will be on the use of the models in question for helping in the analysis of data, not on theoretical derivations.
Prerequisites:
All students will be assumed to be familiar with elenentary statistical concepts, such as are covered in MA25602570. Familiarity with the basic concepts of vectors and matrices is also assumed. (A review of relevant material on vectors and matrices is contained in Appendix B of the text.)
Computing:
Students will be expected to use available computing resources, primarily the Gauss Lab (S110 Ross), or the labs in the Maclaughlin College. Details about computing in these labs (the AML labs) should be obtained from the lab manual handed out in class. Accounts for these labs and the phoenix and Gauss servers can be obtained through MAYA (a.k.a. "Passport York"). For those students who have computers available elsewhere, and need to dial in through a modem, an account on York's highspeed modem pool should also be obtained. (There is a small monthly charge for use of this service.) Printing services are available in the Gauss Lab or in Steacie. There are many statistical programs which could be used for much of the work in this course. We will mostly present sample programs and solutions using SAS, as does the textbook. All assignments, datasets, solutions, hints, other references, test results, etc. will be available only through the course web pages. Students are encouraged to communicate with each other and the instructor via email. To facilitate this, we will make use of a 3330 FORUM, which can be found at this location.
Evaluation:
Students will be expected to turn in assignments roughly bi
weekly. There will be a penalty for late assignments.
There will be 5 roughly biweekly quizzes based on material covered
in the assignments and in class. There will also be a project, due
at the end of term, involving the analysis of a set of data and
the preparation of a report. The practical application of
statistics often involves working as part of a research team, and
so it is recommended that you work in small groups ( absolutely no more than 3 in a group), both for the
project and for assignments. (It is acceptable for groups to hand
in a single assignment, and project.) There will also be a midterm
and a final exam. You will be allowed to use a small set of tables
and notes/formulas for all tests (the exact number of pages to be
detailed in class), but otherwise they will be closed book
affairs. The breakdown of the grading is as follows:
Assignments:  15% 
Project:  05% 
Quizzes:  25% (best 4 out of 5 will count) 
Midterm:  20% 
Final Ex.:  35% 
Outline of coverage from textbook:
Chapters 116, with some omissions of particular topics.
For a schedule of course events please see here.