Key facts
Details for course being taught in current academic year
Level 3 - 15 credits - autumn term
E-learning links
Resources
Course description
Course outline
Full-rank model (multiple and polynomial regression); estimation of parameters, analysis of variance and covariance. Model checking. Comparing models; model selection. Transformation of response and regressor variables. Models of less than full rank (experimental design); analysis of variance, hypothesis testing; contrasts. Simple examples of experimental designs; introduction to factorial experiments. Use of a computer statistical package to analyse real data sets.
Pre-requisite
Statistics and Probability I and II.
Learning outcomes
At the end of this course successful students should be able to:
Understand the theory of the general linear model and derive distributional results relating to
estimators.
Apply the general linear model of full rank to a variety of applications, using transformations and
variable selection techniques if necessary, apply methods for verifying that the model is
reasonable and use the model to answer practical questions, reporting your conclusions in nontechnical
language.
Understand the benefits of designed experiments (including the completely randomised and
randomised blocks designs) and factorial experiments (two factors only), set up and interpret
analysis of variance tables (including linear contrasts), compute confidence intervals, carry out
statistical tests and report conclusions clearly.
Use SAS to fit regression and analysis of variance models, compute additional statistics needed
for applications and construct diagrams required for model verification.
Library
The main recommendation is: Linear Statistical Models, 2nd ed., Bowerman BL and
O’Connell RT, PWS-Kent. This covers nearly all the course, includes SAS output and is not
too expensive. The following two texts cover multiple regression very well but do not have much
on experimental design (and are also very expensive!).
Introduction to Linear Regression Analysis, 2nd ed, Montgomery D C & Peck E A, Wiley
Applied Regression Analysis, 2nd ed, Draper N R and Smith H, Wiley
The following is a very comprehensive book on experimental design.
Design and Analysis of Experiments, D C Montgomery, Wiley.
Two additional books relating to computer software.
Introduction to Regression & Analysis of Variance, Bowman AW & Robinson DR, IOP.
A Handbook of Statistical Analyses using SAS, Everitt B S & Der G, Chapman and Hall.
Assessments
| Type | Timing | Weighting |
|---|---|---|
| Coursework | 20.00% | |
| Practical Report | Autumn Week 7 | 50.00% |
| Project Report | Autumn Week 10 | 50.00% |
| Unseen Examination | Summer Term (2 hours) | 80.00% |
Resit mode of assessment
| Type | Timing | Weighting |
|---|---|---|
| Unseen Examination | (2 hours ) | 100.00% |
Timing
Submission deadlines may vary for different types of assignment/groups of students.
Weighting
Coursework components (if listed) total 100% of the overall coursework weighting value.
Teaching methods
| Term | Method | Duration | Week pattern |
|---|---|---|---|
| Autumn Term | LECTURE | 1 hour | 0101010101 |
| Autumn Term | LECTURE | 1 hour | 2222222222 |
| Autumn Term | PRACTICAL | 1 hour | 1010101010 |
How to read the week pattern
The numbers indicate the weeks of the term and how many events take place each week.