|
Office: |
BSB/202G |
|
Telephone: |
525-9140 x 23423 (24-hour Voice Mail) |
|
e-mail: |
Term 1: Monday 10:30, Wednesday 9:30, Friday 14:30.
Term 2: Tuesday 10:30 & 15:30, Wednesday 10:30, Thursday 9:30
Please come at the start of the hour. Other times by appointment.
Don't hesitate to contact me by telephone, voice mail, or e-mail whenever you need help. If you need to see me at any time and my office door is open, I will see you then if I can, or arrange a time to meet later.
Term 1: Monday 8:30-9:20, Wednesday 8:30-9:20, Thursday 8:30-9:20, in BSB-101.
Term 2: Tuesday 12:30 - 15:30, in BSB-229
The course is intended to ensure a good foundation in the classical theory of statistical inference, and introduce several areas of current research. Some students have more background than others in mathematical statistics and inference, so I will attempt to bring all students up to the same level while keeping the course sufficiently challenging for the more advanced students. Some class time will be used for student presentations of special topics and solutions to assigned problems, to give students experience in presenting and defending their work.
Silvey, S. D. Statistical Inference. Chapman & Hall.
S743 Custom Courseware. (If I get around to preparing it...)
Edwards, A.W.F. (1992) Likelihood, Expanded Edition. Johns Hopkins.
Efron, B. & Tibshirani, R.J. (1993) An Introduction to the Bootstrap. Chapman & Hall.
Hogg, R.V. & Craig, A.T. (1995) Introduction to Mathematical Statistics, Fifth Edition. Maxwell Macmillan. [S3D06 text]
McCullagh, P. & Nelder, J.A. (1989) Generalized Linear Models, Second Edition. Chapman & Hall. [This text is required for S752 in Term II, so it is worth buying it in Term I.]
McLachlan, G.J. & Krishnan, T. (1997) The EM Algorithm and Extensions. Wiley.
Rosner, B.A. (1995) Fundamentals of Biostatistics, Fourth Edition. Duxbury. [S2MA3 text]
Spector, P. (1994) An Introduction to S and S-plus. Duxbury.
In Term I we will spend about 9 weeks working through Chapters 1-7 of Silvey, doing as many of the exercises as possible. Work on likelihood methods will be supplemented by material from Edwards. We will finish Term I by studying the iterative weighted least squares algorithm for fitting generalized linear models, following Chapters 1-2 of McCullagh & Nelder. The emphasis will be on categorical data, with the results applied to biostatistical examples from Chapters 10-11 of Rosner.
Term II will be devoted to special topics of current interest, working from selected journal articles and monograph chapters. Some of the material may be available through the Bookstore as Custom Courseware. The topics may include resampling methods, Bayesian methods, the EM Algorithm, and mixture distributions. Students are welcome to suggest other special topics we might cover in Term II.
Students may use any pocket calculator for assignments, tests and examinations in this course.
For some assignments, a computer may be helpful. Many of the calculations and graphics for this course can be done conveniently on a spreadsheet. A working knowledge of at least one comprehensive statistical package (MINITAB, SAS, SPSS, Splus, Systat, BMDP, etc.) is essential. Students will be encouraged to learn Splus and class examples will be done using Splus. If required, class time can be allocated to computer tutorial sessions. Specialized software, such as GLMStat for generalized linear models, will also be used. Familiarity with BASIC, FORTRAN or another programming language will also be useful, for calculating non-standard methods not available in packages.
There will be a take-home written examination in May 1999, based on all the material covered in the course. Students may use computers for the examination. Students may approach the instructor for clarification of any questions but they must work independently of each other. Students may not discuss the questions with anyone other than the instructor. The paper will be handed out on the morning of 1999-05-03 and all work must be submitted by 1999-05-14 17:00.
It is each student's responsibility to keep up to date with the course by working ahead on examples and exercises from the texts and reference material.
Five assignments will be given. To keep up to date with the course, it will be essential to complete and submit each assignment on time.
If time permits, students may have opportunities in class to present their solutions to assigned problems and introduce selected topics. Students may be asked to submit their presentation notes in legible form for evaluation.
All assignments will be counted. Class presentations and participation will be evaluated.
50% Final Examination + 40% Assignments + 10% Class Presentations and Participation;
I will review all "borderline" marks and possibly make further adjustments.
We remind you of the "Statement on Academic Ethics" and the "Senate resolution on Academic Dishonesty" as found in the Senate Policy Statements distributed at registration and available in the Senate Office. Any student who infringes one of these resolutions will be treated according to published policy.