MSc in Statistics
A flexible taught masters course in statistics combining in-depth training in mainstream advanced statistical modelling with a broad range of specialisations - from financial mathematics to statistical bioinformatics; from shape analysis to risk management.

Commencing: September 2011

How to Apply:
You can apply online by visiting this weblink:
You can also download a form for completion by hand from this link.

Want to know more?
Continue reading this page, or contact

Ms Iwona Malinowska
Taught Postgraduate Administrator
School of Mathematics
Telephone: + 44(0) 113 343 5111

Dr Stuart Barber
Programme manager, MSc Statistics
Telephone: +44 (0) 113 343 5146

Why study for an MSc in Statistics?

There is a shortage of well-qualified statisticians in the UK and other countries. Numeracy, in general, is an attribute keenly sought after by employers.

The emergence of data mining and analysis means that demand for statisticians is growing across a wide range of professions - actuarial, betting and gaming industries, charitable organizations, commercial, environmental, financial, forensic and police investigation, government, market research, medical and pharmaceutical organisations. The course is designed specifically to meet this demand.

Many statistical careers require people educated to masters degree level. This course is designed to build on existing mathematical skills and deepen knowledge of statistics in order for you to access a variety of professions or pursue further research as a PhD student.

The MSc is accredited by the Royal Statistical Society. Completion of the programme will qualify students for the status of ?Graduate Statistician? - the first stage towards Chartered Statistician status.

The School of Mathematics

The School comprises three Departments: Applied Mathematics, Pure Mathematics, and Statistics. Together, they form one of the largest research schools of mathematicians in the UK, with around 65 academic staff, 20 postdoctoral fellows and 75 PhD students.

Each Department has an enviable research reputation - each one rated 5 ("internationally excellent") in the 2001 UK Research Assessment Exercise. In the recent 2008 RAE, the Statistics Department had 65% of its research graded as "world leading" or "internationally excellent".

About the course


MSc in Statistics

The MSc in Statistics at the University of Leeds is a flexible degree programme enabling students from a wide range of backgrounds to both broaden and deepen their understanding of statistics.

The programme provides training in a range of statistical techniques (and transferable skills) suitable for either careers in statistics and research-related professions, or for further academic research in statistics.

Options within the course vary from mainstream topics in statistical methodology to more specialized areas and reflects specific research interests of academic staff within the Department - examples include statistical shape analysis, directional data, statistical genetics and stochastic financial modelling.

The course is a full-time MSc lasting 12 months, starting in September. The Diploma is just over 8 months duration.

The Course consists of two semesters of taught modules, with the third semester (MSc only) devoted to a major dissertation. Within each semester there is one compulsory module. However, students tailor the course to meet their individual needs through selection of a further three modules from a variety of options. This selection process means a student can specialise in biological or financial applications of statistics or retain a broad base of statistical expertise.

Why study Statistics at the University of Leeds?

Simply because you will be taught by leading figures in their field of specialisation! We have recently made several exciting new appointments including Professor Wally Gilks to head up our new Centre for Statistical Bioinformatics.
Our School?s reputation is recognised worldwide through research collaborations with other universities, with government bodies, hospitals, sectors of industry and through a variety of interdisciplinary projects with other departments within the University.

Our Department is home to LASR - Leeds Annual Statistics Research workshop - an event originated (and still led) by Professor Kanti Mardia over 25 years ago. It has grown from a small group of specialists sharing common interests in development of their research into an international platform attracting experts at the cutting-edge of statistical research including spatial statistics; shape analysis, bioinformatics and image analysis.

Other areas of research include:

Modern Data Analysis including robustness (a key concept in modern statistical modelling); pattern recognition and machine learning; statistical computation and wavelets.

Statistical Bioinformatics including phylogeny; sequential analysis; structure analysis and gene expression analysis.

Probability and Stochastic Processes comprising parameter estimation for Markov processes; stochastic financial models; stochastic differential equations; propagation and extinction in catalytic media; intermittency in random media; limit theorems for sums and extreme values and probabilistic methods in number theory and combinatorics.

The Course

A formal description of the course structure and options can be found in the MSc Statistics entry in the University's Programme catalogue. Here is a brief summary.

Semester One

Statistical Computing (compulsory)
An introduction to methods of statistical computing. Essential for the applied statistician, with an emphasis on sampling-based methods, such as Markov chain Monte Carlo.

Optional Modules
  • Multivariate and Cluster Analysis
  • Statistics and DNA
  • Linear Regression and Robustness and Smoothing
  • Stochastic Financial Modelling
  • Discrete Time Finance
  • Core Epidemiology
  • Advanced Epidemiology

Semester Two

Independent Learning Skills (compulsory)
An introduction to research methods including literature search, writing styles, mathematical typesetting and programming skills.

Optional Modules
  • Statistical Theory
  • Times Series and Spectral Analysis
  • Generalised Linear and Additive Models
  • Continuous Time Finance
  • Risk Management
  • Hidden Markov Models
  • Multilevel and Latent variable Modelling
  • Advanced Modelling

Semester Three

Dissertation in Statistics
A four-month research project undertaken in the Summer on a topic (chosen in conjunction with project supervisors) culminating in a dissertation on that project.

Example Project Topics include:
  • Shape Analysis
  • Spatial Statistics
  • Robustness
  • Data Mining
  • Wavelets
  • Bioinformatics
The taught course is primarily assessed by end-of-semester examinations with a small component of continuous assessment. The Semester Three project is assessed by a written dissertation and a short oral presentation.

Optional Module Descriptions

Multivariate and Cluster Analysis
Some common methods, like t-tests, can be extended to the multivariate setting and many new approaches are now available. Principal components analysis and factor analysis look for ?interesting? combinations of the variables; discriminant analysis tries to separate different groups of data; cluster analysis tries to find natural groupings in the data.

Statistics and DNA
Modern biological experiments produce large data sets involving information related to DNA. This module gives the basic biological background before looking at a range of data types and methods to analyse them.

Linear Regression and Robustness and Smoothing
A fundamental statistical tool is the simple linear regression model, which predicts the value of a normally-distributed response variable from a predictor variable. This module explores many ways to extend this simple model to cope with non-linear relationships and data corrupted by non-normal errors.

Stochastic Financial Modelling
Financial investments such as stocks and shares are risky: their value can go down as well as up. To compensate for the risk in a fair market, a discount is needed. This module will develop the necessary probablistic tools to enable investors to value such assets.

Discrete Time Finance
This module develops a general methodology for the pricing of financial assets in risky financial markets based on discrete time models.

Shape Analysis and Directional Data
The "shape" of an object is the information that doesn't change when the object is moved, rotated, enlarged or reduced. Specialised statistical methods are needed to study shape variation.

Statistical Theory
We often use statistical tests and estimators without fully exploring the theoretical basis for their use. Here, we look more deeply into the mathematics behind statistical inference and compare the two main approaches to statistics: Frequentist and Bayesian inference.

Time Series and Spectral Analysis
In time series, measurements are made at a succession of times, and it is the dependence between measurements taken at different times which is important. We concentrate on techniques for model identification, parameter estimation and forecasting future values of the time series.

Continuous Time Finance
Continuous time models play a central role in pricing of financial assets under more challenging circumstances than can be handled with discrete time models.

Risk Management
This module gives comprehensive coverage of mathematical and practical approaches to financial risk management. Avoiding the disastrous consequences of badly managed risk requires detailed mathematical knowledge of how to quantify financial risk and stress-test a hedge.

Other Information

Opportunities for postgraduate research in the Department of Statistics

Department of Statistics web pages

University of Leeds | Last modified: Mon Aug 16 10:57:49 BST 2010 | Privacy Statement