I currently teach two modules.
I teach the first half of this module, in which we consider mixed effects models, which are a way of introducing hierarchical structure into linear models. By grouping subjects at various levels, we can hope to achieve more realistic uncertainty estimates in our predictions. The module relies heavily on the lmer package in R, and I introduce the use of JuPyTeR notebooks for doing reproduceable open data analysis.
Modern computational tools for the implementation of the frequentist and likelihood-based approaches to inference are explored, with strong emphasis placed on the use of simulation and Monte Carlo methods. Frequentist proceedures such as hypothesis testing make most sense to me when viewed as repeatedly sampling from various hypotheses. Indeed, many of the people who derived these tests, such as Fisher, viewed them as approximations to what they would do if had the ability to repeatedly compute. We now have that ability, and so you can forget all of the sampling distributions associated to the multiple different hypothesis tests learnt previously.
I taught three different modules at the University of Nottingham
Statistical methods and models (G12SMM) I taught the linear models part of the module, covering theory and application. I created a set of R markdown case studies for demonstrating various aspects of the analysis. You can find these on my github page.
Topics in Biomedical Statistics (G14TBS) This is a level four/Msc module which takes place in the Spring Semester. I taught and wrote the part on population genetics, which focusses on explaining how probability models can be used to understand the genetic structure of populations. The notes are available here.
Computational Statistics (G14CST) A fourth year module on modern computational techniques needed by most practicing statisticians.
I’ve given a number of research talks that aim to be tutorials.
GP summer schools NIPS