|
Module Availability |
Autumn semester
|
|
|
Assessment Pattern |
Unit(s) of Assessment
|
Weighting Towards Module Mark( %)
|
2 hour unseen examination
|
75%
|
Test
|
10%
|
Coursework
|
15%
|
Qualifying Condition(s)
A weighted aggregate mark of 40% is required to pass the module.
|
|
|
|
Module Overview |
Fundamental topics in the design and analysis of experiments are introduced. For a variety of statistical models, the structure of the model and applications are covered. Particular attention is given to model adequacy checking. Statistical software is used to ensure that the emphasis is on methodological considerations rather than on calculation. |
|
|
Prerequisites/Co-requisites |
MAT2002 General Linear Models would be useful but is not a pre-requisite. |
|
|
Module Aims |
The principal aims are to expose students to a wide range of statistical designs and concepts and to provide them with the tools to design and analyse appropriate experiments in a range of situations.
|
|
|
Learning Outcomes |
At the end of the module a student should have:
(1)an appreciation of design consideration and constraints
(2)an ability to design and analyse an experiment in a variety of situations.
|
|
|
Module Content |
General Concepts:
Principles of design and strategy of experimentation Complete designs: m-way classification
Designs Involving Blocking:
Precision improvement by blocking
Randomized block designs
Incomplete block designs and balance
Row column designs
Euler's conjecture: Graeco-Latin squares Youden squares
Further Topics Involving Blocking:
Efficiency
Optimality criteria
Connectivity
Factorial Designs:
Principles and advantages of factorials
Two level factorial systems
Fractional factorials
Confounding in factorials with n factors each at 2 levels
Topics with Specific Applications:
Robust design and Taguchi methods
Analysis of covariance
Binary response data
Crossover designs and carryover effects |
|
|
Methods of Teaching/Learning |
Teaching is by lectures and example classes. Learning takes place through lectures, exercises (example sheets) and background reading.
Autumn semester: 3 contact hours per week for 10 weeks. Mainly lectures but including some supervised computer lab sessions. |
|
|
Selected Texts/Journals |
Angela Dean and Daniel Voss, Design and Analysis of Experiments, Springer, 1999
J.A. John and E.R. Williams, Cyclic and Computer Generated Designs, Chapman and Hall, 1995 Douglas C. Montgomery, Design and Analysis of Experiments, 6th edition, Wiley, 2004
|
|
|
Last Updated |
04.11.08 |
|