Statistics and applied probability

Statistical methodology and development of new probabilistic techniques inspired by applications including research in stochastic and probabilistic modelling and inference in stochastic systems.

This is an area of strength for the UK and importance for many scientific disciplines. Despite substantial growth over the last Delivery Plan, demand is undiminished for qualified statisticians with an understanding of application areas including data analytics, healthcare modelling and Artificial Intelligence. Statistics and Applied Probability research has major impacts in many areas, as evidenced by its contribution to seven EPSRC Prosperous Nation Ambitions. It is a major contributor to advances in data science, healthcare and the digital economy. 

We will develop a focus on statistics, in order to balance the research area. We will encourage applications with a high proportion of fundamental statistics; this can be coupled with: applied probability, fundamentals for AI, big data or model development but the majority of the work must fall under the remit of statistics.

This strategy aims to build on recent investment in the area (e.g. the Alan Turing Institute), address capacity issues throughout the 'people pipeline' and respond to growing demand across all sectors of economy and society by growing investment and supporting people at all career stages within this research area.

By the end of the current Delivery Plan, we aim to have:

  • Supported research and training that builds on and complements previous and current work, including activities by the Alan Turing Institute. This will mean maintaining support for core fundamental statistical methodologies, while developing links with more applied areas of statistics across the entire research landscape, both within EPSRC's domains and more broadly. Outputs from the Statistics and Applied Probability Landscape Event (STAPLE) will be used to develop a highly collaborative portfolio.
  • Supported research aligned to the Global Challenges Research Fund (GCRF), e.g. in the areas of uncertainty and epidemiology.
  • Responded to the growing demand for people with skills in statistics and applied probability, particularly at the early-career stage. It is important to ensure that people have skills across all areas of statistics and applied probability, as well as spanning key topics such as machine learning, data analytics, uncertainty quantification and medical statistics.
  • Focus fellowship priority areas for Statistics, with applications needing to contain at least 60% Statistical research in order to be within remit.

The UK has an international reputation for expertise in a number of statistical methods, including medical statistics, bayesian statistics, interface with genomics, machine learning and big data. There are strong connections between Statistics and Applied Probability and an array of applications in sciences, industry, business and government - providing economic, industrial and societal impact in a range of applications and sectors (e.g. healthcare and modelling financial and environmental uncertainty) (Evidence source 1,2,3,4,5,6,7).

Statistics and Applied Probability is therefore an important research area that connects to and supports a number of other research areas, key topics and disciplines (e.g. healthcare, data analytics and uncertainty quantification) (Evidence source 1-9). This importance is reflected in the 2012 Deloitte report, the 2014 and 2016 EPSRC Statistics and Applied Probability theme days, the 2014 Statistics Strategy from the Office for National Statistics and the area’s relevance to large investments including the Alan Turing Institute (Evidence source 1,2,3,4,6,7). Over half of EPSRC investments in this area are relevant to industrial sectors such as healthcare, environment, financial services and energy. The research area is also relevant to the 'eight great technologies', primarily big data and robotics and autonomous systems (Evidence source 1,8).

Statistics and Applied Probability has links to many research areas and Themes across EPSRC - most notably within Information and Communication Technologies (ICT), maths, healthcare technologies and digital economies. Within the Mathematical Sciences Theme, the primary connections are with Numerical Analysis, Non-Linear Systems and Mathematical Analysis; across EPSRC, the primary connections are with Artificial Intelligence Technologies, Healthcare Technologies and Manufacturing the Future Themes.

Statistics and Applied Probability researchers have a broad range of skills, including modelling, optimisation techniques, uncertainty quantification, data analytics and machine learning. There is demand from industry to recruit researchers with this knowledge of fundamental statistical and probabilistic methodology, but a recognised shortage of these skills in the UK is a concern.

Over the last Delivery Plan, capacity has grown at all career stages and recruitment at PhD level is healthy, with three Centres for Doctoral Training (CDTs) supported in this area. But there is still significant concern about recruitment and retention of skilled academics and the threat of key capacity being lost from UK academia to industry, with universities being unable to compete with industry. In this regard, academia has the opportunity to complement industry's research interests rather than trying to replicate them, and there is a need to support career progression at all stages - especially the early academic career stage (Evidence source 1,2,3,6,7).

This area is of substantial relevance to all Outcomes, with short, medium and long-term contributions. Ambitions where this area contributes most significantly to the Connected, Healthy, Prosperous and Resilient Nation Outcomes are:

C1: Enable a competitive, data-driven economy

Statistics will be a key contributor in terms of using novel mathematics and statistics techniques and translating these skills to realise the benefit to business through forecasting and decision-making. Statistics supporting machine learning (e.g. development of online change-point methods) is key.  

C3: Deliver intelligent technologies and systems

New technologies will include supporting decision-making and using data for application.

H1: Transform community health and care

The need for real-time information and development of models highlights the importance of statistics in ensuring how data is used to develop reliable models.

H3: Optimise diagnosis and treatment

There is a need to develop optimised models, particularly accounting for the statistical modelling of uncertainty.

H4: Develop future therapeutic technologies

Statistics is expected to make an important contribution, particularly in the area of adopting efficient clinical trials.

P4: Drive business innovation through digital transformation

This area has applications in intelligent technologies and data analytics.

R3: Develop better solutions to acute threats: cyber, defence, financial and health

There is a need for modelling of uncertainty, quantifying this and providing predictions and decisions using historic data and models.

Research area connections

This diagram shows the top 10 connections between Research Areas within the EPSRC research portfolio. The depth of the segment relates to value of grants and the width of the segment relates to the number of grants shared by those two Research Areas. Please click to see the related Research Area rationale.

Visualising our Portfolio (VoP)
Visualising our portfolio (VoP) is a tool for users to visually interact with the EPSRC portfolio and data relationships.

EPSRC support by research area in Statistics and applied probability (GoW)
Search EPSRC's research and training grants.

Contact Details

In the following table, contact information relevant to the page. The first column is for visual reference only. Data is in the right column.

Name: Laura McDonnell
Job title: Portfolio Manager
Section / Team: Mathematical Sciences
Organisation: EPSRC
Telephone: 01793 444268