Math 356H: Linear Statistical Models

Instructor: Michelle Boué
Office Phone:748-1011 x7925
Office: Peter Gzowski College 332 (Enweying)
Secretary:Carolyn Johns
Office Phone:748-1011 x7531
Fax:748-1011 x1555
Office:Peter Gzowski College 342 (Enweying)

MATH 356H course materials for the Winter 2008 term can be found

Course Topics and Objectives

Mathematics 356H provides an introduction to the study of linear statistical models for regression, analysis of variance and experimental designs. Extensive use of statistical software is made throughout the course.


Required: Probability and Statistics for Engineering and the Sciences, by Jay L Devore (Fifth or Sixth Edition), Duxbury.

Further reading:

  • Neter, J. Wasserman, W., Kutner, M.H. (1990), Applied Linear Statistical Models, Irwin.
  • Draper, N., Smith, H. (1981) Applied Regression Analysis, Wiley.


Any material relevant to the course as well as updated grades will be available on


This course is intended for students who have completed MATH 256H. Previous specific computing experience is not required for those parts of the course involving computer-based analysis.


Due to the considerable amount of numerical work involved in this course, students should possess a calculator with built-in statistical function keys.

Course Schedule

There will be three lecture hours per week as indicated in the Class timetable. Students are responsible for all material covered in lectures and for all announcements made in lecture hours.

Course Evaluation

Problem Sets: There will be five problem sets through the year. Each problem set will contribute 8% of the final mark.
Midterm examinations: There will be two in-class exams. Each exam will contribute 15% of the final mark.
Final Examination: There will be a final three-hour examination during the final examination period. The final examination will contribute 30% of the final mark.
Problem sets:     5 x 8%         40%
Midterm exams     2 x 15%         30%
Final exam     1 x 30%         30%

Course Outline

  1. Simple Linear Regression
    1. The simple linear regression model
    2. Least squares estimation of the regression parameters
    3. Inferences
      1. Inferences concerning the slope parameter
      2. Inferences concerning the intercept
      3. Interval estimation of the mean
      4. Prediction of new observations
    4. Correlation
    5. Remedial measures
  2. Matrix Approach to Linear Regression
  3. General Regression Models
    1. Polynomial regression
    2. General multiple regression
    3. Diagnostics and remedial measures
    4. Building the regression model
  4. Analysis of Variance
    1. Single-factor ANOVA
    2. Multiple comparisons
    3. Multifactor ANOVA
    4. Non-parametric approach
  5. Experimental Designs
    1. Randomized block designs
    2. Latin squares

Academic Dishonesty

Academic dishonesty, which includes plagiarism and cheating, is an extremely serious academic offense and carries penalties varying from failure in an assignment to suspension from the University. Definitions, penalties, and procedures for dealing with plagiarism and cheating are set out in Trent Universitys Academic Dishonesty Policy which is printed in the University Calendar.

Access to Instruction

It is Trent University's intent to create an inclusive learning environment. If a student has a disability and/or healh consideration and feels that he/she may need accommodations to succeed in this course, the student should contact the Disability Services Office (BL Suite 109, 748-1281, as soon as possible. Complete text can be found under Access to Instruction in the Academic Calendar.