Handbook

This unit explores the statistical modelling methods that underlie the analytic aspects of Data Science and Machine Learning. By working through examples, this unit gives a strong mathematical and statistical foundation to enable a deeper understanding of data analysis and machine learning methods taught in later MDS/MAI units which focus on machine learning with a more practical perspective. It introduces basic notions about data and foundational mathematics and statistics in the form of sample statistics, probability, expectation and parametrised probability distributions. This provides a basis to introduce statistical inference through maximum likelihood estimation, confidence intervals and hypothesis testing as a way of inferring information about the probability distributions that best describe observed data. Building upon inference models, the unit considers predictive models that predict one data variable based on other data variables through introductory supervised machine learning methods for regression and classification. Unsupervised machine learning methods such as clustering that find hidden groupings in data are also considered.