MSc and PG Diploma in Applied Statistics
The MSc in Applied Statistics is a 12-month taught Masters degree running from October to September each academic year. The MSc in Applied Statistics trains you to solve real-world statistical problems. Faced with a problem of data analysis, graduates know how to choose and adapt appropriate statistical and computational methods, implement the analysis on a computer and communicate results clearly and succinctly.
The MSc has a particular focus on modern computationally-intensive theory and methods. It provides a broad high-level training in Statistical Machine Learning, Applied and Computational Statistics and the fundamental principles of Statistical Inference. Training is delivered through mathematically demanding lectures and problems classes, hands on practical sessions in the computer laboratory, report writing and dissertation supervision. You will have around three months to work on your dissertation with guidance from your supervisor.
You will be assessed on your performance in written examinations around May, through your work in assessed practical problems set throughout the year, and by the quality and depth of your dissertation. The MSc dissertation is completed by mid-September.
The Postgraduate Diploma in Applied Statistics is a 9 month course, starting in October and ending in June. It is similar to the MSc but shorter as no project is required.
Our new delivery and content
We are making some changes to the content and delivery of the MSc in Applied Statistics. The new programme will be delivered for the first time in 2016-2017. We have more material on Statistical Machine Learning and Computational Statistics and the number of lectures in options courses is set to increase as we open up our third and fourth year undergraduate courses to MSc students. The majority of lectures in the new MSc will be delivered in courses alongside senior undergraduates.
Students take four or exceptionally five courses each term. All courses are sixteen lectures. Three courses each term are core courses, and students must complete the practical sessions in these courses.
Applied Statistics (theory and application of linear and mixed models and generalised linear models).
Statistical Inference (Bayes & Frequentist estimation; Decision theory; Variational Methods and EM).
Statistical Programming (MSc only - R programming; Graphics and visualisation; Advanced R).
Computational Statistics (Non-linear and non-parametric models; Bootstrap; Hidden Markov Model).
Bayes Methods (Prior elicitation; Bayesian non-parametrics; Approximation methods; Case studies).
Data Mining and Machine Learning (Unsupervised Machine Learning; Kernel and Ensemble methods).
Options will vary from year to year. In 2016-2017 we expect the first term options to be:
Probability and Statistics for Networks
Mathematical Genetics (although possibly Mathematical Genetics will be taught in the second term rather than the first).
We expect the second term options to be:
Statistical Machine Learning
possibly plus Advanced Simulation (the Advanced Simulation course will run if departmental resources allow and will not otherwise)
Changes to course for 2016/2017 including course descriptions [PDF]
- Graduate admissions forms and procedures
- Graduate prospectus entry
- Graduate funding pages
- Frequently asked questions about the admissions process (FAQs)
- Find out more about the course