An individual assignment weighted at 40% exploring a variety of experimental designs and performing appropriate analysis for a range of problems. Use of statistical software is also required. (Learning outcomes 1, 2)

An Exam length 2 hours weighted at 60% covering the statistical content of the module (Learning outcomes 3 and 4)

The exam will be the final assessment point.

Prior study of MATH50400 Survey Design and Statistical Inference or equivalent

SPSS and R

For background reading only, e.g.
An Introduction to the Design and Analysis of Experiments, George Canavo and Ioannis Koutrouvelis, Pearson 2009, ISBN: 0136158633

Discovering Statistics using SPSS, Andy Field, Sage 2009, ISBN: 9781849204088

Forecasting Principles and Applications, Stephen A. DeLurgio, McGraw-Hill, 1998, ISNB: 0256134332

An Introduction to Generalised Linear Models, Dobson, A. J. & Barnett, A., CRC Press, 2011, ISBN: 9781584889502

Students are required to commit to 300 learning hours of which 60 hours will consist of contact time. Typically there will be 24 hours of lectures (1 a week) and 36 hours of tutorial/practical time (alternating 1 and 2 a week). Lectures will provide students with a broad overview of the indicative content and theoretical concepts. They will apply these concepts to real data sets using an appropriate statistics package. The tutorial/practical sessions will also allow students to practise the material covered in the lectures through problem solving.

Design and implement experiments involving: Completely Randomised Designs;
Randomised Block Designs; Latin Square Designs, Incomplete Blocks.
General ANOVA experiments and dealing with missing values.
Factorial Experiments
Non Parametric methods common in the analysis of experiments.
Multiple Linear Regression, weighted least squares regression.
Binary regression, logistic regression models, odds ratio, use of stepwise regression
Generalized linear models, Poisson regression, Binomial regression
Loglinear models for contingency tables
Time series analysis
Introduction to multivarate techniques

1) Construct an experimental design to satisfy given requirements. (Communication, Knowledge and Understanding, Problem Solving).
2) Analyse experimental data using appropriate statistical techniques, interpret the results and reflect on the validity of the output from a statistical package. (Analysis, Application, Knowledge and Understanding, Reflection).
3) Understanding the principles associated with statistical modelling. (Learning, Knowledge and Understanding).
4) Use and interpret the output of a statistical package to formulate a statistical model to appropriate situations - checking assumptions where necessary. (Analysis, Application, Enquiry, Learning).