MATHS 429: Analysis of Variance in Experimental Design Models (3)

Syllabus

 

1.  Prerequisite:  MATHS 321 or equivalent.

2.  Course Description:  Multivariate normal distribution; quadratic forms; linear models; simple random, randomized block, Latin squares, factorial, split-plot, balanced incomplete block design; analysis of covariance; confounding; multiple comparison tests.

3.  Course Objectives:  Experimentation is a key component in research.  In the initial phases of developing a research program, the various aspects of experimental design should be incorporated. The initial research proposal should address such issues as: which experimental setup to use, which factors to study, the manner of obtaining the observations, and so on.  Aspects of experimental analysis are used in preparing final reports of the work, and in the dissemination of the research results. Extracting relevant information from the data, constructing valid estimates, and verifying the existence of meaningful differences are some of these aspects.  This course considers both the design and analysis phases of experimentation.

4.  Course Rationale:  The role of a statistician in an experimental program is not merely to analyze the results of the experiment, but also to help the scientist plan his experiment in such a way as to produce valid results as efficiently as possible.  This course would prepare a student for such a challenge.  The course is a required course for students majoring in statistics and is also an excellent course for students who are in mathematics, business, psychology, and a host of other areas where statistical methods are employed.

5.  Course Content: Multivariate normal distribution. Quadratic forms. Linear models. Designs: simple random, randomized block, Latin squares, factorial, split plot, balanced incomplete block. Analysis of covariance. Confounding. Multiple comparison tests.

6.  Course Format:  The class will be taught through a lecture/discussion format. Students will be expected to participate in class regularly, to study the appropriate material, and to complete a variety of assignments using statistical software.

7.  Evaluation of Student Performance: Course grades are determined primarily by student performance on tests, quizzes, and projects, as well as possibly homework and class participation.  The evaluation and weight of these various components are at the discretion of the individual instructor.

8.  Evaluation of the Course:  The instruction of the course is evaluated by departmental student evaluations and peer evaluations.  The course is reviewed and revised periodically by the Department Undergraduate Programs Committee.

[VDM 10/2005]