This unit covers a set of tools for modelling, understanding and predicting from complex data sets. The tools are selected from topics that are a natural blend of statistics and machine learning, and are motivated and demonstrated with applied examples. The underlying general concepts and basic theory are discussed at a level accessible to students. Data sets are analysed using the statistical package R and the unit provides an introduction to this software. Topics are selected from statistical inference, linear regression, model selection, classification, resampling methods, tree-based methods, support vector machines and machine learning.

Students are able to (1) apply appropriate techniques from the above topics to real world data and communicate results in a logical and coherent fashion; (2) apply statistical reasoning in general to analyse the essential structure of problems in various fields of data science; (3) extend students' knowledge of statistical modelling techniques and adapt known solutions to different situations; and (4) undertake continuous learning in statistical predictive modelling and inference, being aware that an understanding of fundamentals is necessary for effective application.