This 5-day course will introduce participants to a range of multivariate techniques that are becoming more commonly used in the analyses on large and increasingly complex observational epidemiological data sets. The techniques covered include: cluster analysis (hierarchical and partitional), multi-dimensional scaling and other network plotting methods, dimension reduction techniques (PCA, common factor analysis, and MCA), as well as a variety of tree-based and other approaches to classification in high-dimensional space. For all of these approaches, a strong emphasis is being put on the use of visualisation both to explore the patterns and associations within the data and also to aid in the interpretation of the modelled outputs. During practicals, participants are given the opportunity to apply these methods to course data sets as well as any own data brought to the course. We will use the Python programming and data analysis platform to introduce and work with the methods, with corresponding R code supplied for the main portions of the course.
- This event has passed.