The course begins with an advanced introduction to probability, statistics and data analysis, including testing, estimation, linear regression, model selection and logistic regression. This lays the foundation for concepts of modern predictive modelling and machine learning such as predictive error, loss functions, overfitting, generalisation, regularisation, sparsity. Techniques include modified regression, recursive partitioning, boosting, neural networks. Application to real data sets from a variety of sources, including data preparation and checking, model selection and evaluation, and reporting.