Knowledge of statistical modelling is useful in many graduate careers. This double module gives students a practical understanding of various statistical modelling techniques that are commonly used by professional statisticians and is intended to prepare students for a career involving statistical analysis.

The module begins with the study of non-parametric methods for statistical inference and statistical modelling, including methods based on resampling and order statistics. This is followed by an exposition of multivariate linear regression, including some discussion of ridge regression and experimental design. We conclude with a study of the Generalized Linear Model (GLM) and derive logistic regression and ANOVA as special cases.

Prerequisite Module: MA2502 Computational Statistics

Recommended Module: MA2500 Foundations of Probability and Statistic

• Understand the principles and applications of non-parametric methods.
• Explain the theoretical foundations of multivariate regression.
• Apply the GLM in different contexts

Modules will be delivered through blended learning. You will be guided through learning activities appropriate to your module, which may include:

• Weekly face to face classes (lectures, labs and exercise classes)
• Electronic resources that you work through at your own pace.

Students are also expected to undertake self-guided study throughout the duration of the modul

Skills:

An ability to choose and apply appropriate statistical modelling techniques in various practical settings, to quantify the uncertainty associated with various statistical models, and to analyse, interpret and present the results of statistical analysis.

Transferable Skills:

Formulating problems and presenting clear solutions through logical reasoning.

• Non-parametric methods
• Linear regression
• The Generalized Linear Mode