Autumn

Unit(s) of Assessment	Weighting Towards Module Mark( %)
2 hour unseen examination	75%
Coursework	25%
Qualifying Condition(s)   A weighted aggregate mark of 40% is required to pass the module.

This module introduces least square fitting, methods of inference based on normal theory, diagnostic and analysis if data from simple designs.

MS132 Probablilty and Statistic (MAT1017 Proof Probability and Experiment)

The aims of the module are to introduce concepts involved in general linear models and to equip students with the diagnostic techniques necessary to assess the suitability of a given model. The methods used in analysing simple one-way and two-way experiments and Latin squares designs are also covered.

At the end of the module a student should:

• be familiar with the main results and methods of the linear and generalised linear models considered in the module;
• be able to apply these results to analyse appropriate data;
• be able to interpret the results from such analyses.

• Review of one- and two sample normal-based methods, revision of R. Covariance and correlation. The linear model. Least squares estimation. Simple and multiple regression. Selection of variables.
• Completely randomised and randomised block experiments. One- and two-way analysis of variance. Interaction. Contrasts. Latin square designs and binary data. General regression approach to analysis, residual analysis and diagnostics.

Teaching is by lectures, tutorials and computing labs. Learning takes place through lectures, tutorials, practicals, exercises and background reading.

3 contact hours per week for 10 weeks. Mainly lectures/tutorials, but including supervised lab sessions.

D. C. Montgomery, Design and Analysis of Experiments, Chapman and Hall, 1993.

W. J. Krzanowski, An introduction to statistical modelling, Arnold , 1998.

B. S. Everitt, Statistical Analyses using S-Plus, Chapman and Hall, 1994.

G. B. Wetherill, Intermediate Statistical Methods, Chapman and Hall, 1998.