Departmental news

The Statistics Department has recently been awarded funding from NERC for the project "Statistical inference and uncertainty quantification for complex process-based models using multiple data sets". Principal Investigator Richard Everitt, will collaborate on the project with other members of the Department (Rito Dutta, Christian Robert and Martyn Plummer), the Ecology group at the University of Reading, and with the Centre for Environment, Fisheries and Aquaculture Science.

Making responsible decisions about landscapes is facilitated by the use of complex models able to represent multiple competing demands on land use. Decisions about land use require that trade-offs between competing demands be identified, and their consequences through time be characterised. Models consisting of stochastic computer simulations are increasingly used to make realistic predictions about real world processes from socio-ecological systems involving land use. dels attempt to simulate all relevant aspects of a real physical system, they may involve many parameters, some of which will be difficult to set correctly. The final objective of these models is to assess the possible consequences of management decisions, such as the placement of wind turbines, thus it is crucially important that the uncertainty introduced by calibrating parameter values be understood.

In order to make informed decisions, one needs to be able to consider the effects of a number of complex interacting temporal and spatial processes (e.g. hydrological, ecological, agricultural, economic, climate). The project will develop new techniques in Approximate Bayesian Computation to enable parameter estimation for models for these processes, taking into account the impact of model misspecification. This project is part of the Strategic Priorities Fund on Landscape Decisions. https://landscapedecisions.org/

Chenlei Leng, Professor of Statistics, University of Warwick, has been named Fellow of the Institute of Mathematical Sciences (IMS). Professor Leng received the award for fundamental contributions to the theory and practice of high dimensional statistics, statistical machine learning, model selection and network data analysis.

Created in 1935, the Institute of Mathematical Sciences is a member organisation that fosters the development and dissemination of the theory and applications of statistics and probability. The IMS has 3,500 active members throughout the world.

https://imstat.org/2020/05/19/ims-names-2020-fellows/

Professor Xavier Didelot is a member of a team of Warwick researchers awarded a £1m grant from the Wellcome Trust to work with East African countries in their emergency preparations for COVID-19 as the pandemic spreads across Africa.

https://warwick.ac.uk/newsandevents/pressreleases/warwick_researchers_to/

Congratulations to Tom Berrett on being awarded the Royal Statistical Society Research Prize . The citation notes his outstanding contributions to understanding and developing nearest neighbour methods for classification, entropy and related functional estimation, and his highly original work on independence and conditional independence testing. Tom will join the Department of Statistics in July.
https://www.statslife.org.uk/news/4453-announcing-our-honours-recipients-for-2020

The £4 million grant from NIHR is part of an announced £58.7 million research investment to protect the public from health threats such as antimicrobial resistance, air pollution and infectious diseases. The HPRU in Genomics and Enabling Data will ensure that cutting edge genomic methods are being used to protect public health.

In partnership with researchers from Public Health England, the Wellcome Sanger Institute and the University of Cambridge, Professor Didelot will lead a team of researchers from the School of Life Sciences, Mathematics Institute, Department of Statistics and Warwick Medical School. The research aims to reduce the burden of infectious diseases, investigate the likely effects of control strategies, and make sure that healthcare resources, especially antibiotics, are used optimally. Over the next five years, scientific researchers from all four institutions will use their expertise to look at the genomes of infectious diseases with one collective data bank to improve English public health.

Professor Didelot comments:

“It is a great pleasure and privilege to have been selected to lead this Health Protection Research Unit. I am looking forward to working closely with all partners at the University of Warwick, Public Health England, the University of Cambridge and the Wellcome Sanger Institute.
This grant from NIHR means that over the next five years we will work together to develop the use of new genomic methods to protect and improve the nation's public health, including looking at the detection and spread of diseases, or tackling antibiotic resistance.”

Launching this week for the summer of 2020; an exciting new 8 week long research internship program, run by the Department of Statistics. Projects across the fields of Statistics, Probability and AI aimed at giving successful applicants an ideal introduction to a career in research.

Candidates will be guided by an academic supervisor and they will also have the opportunity to participate in weekly cohort events. More information, including details of how to apply, can be found at http://warwick.ac.uk/stats-interns. DEADLINE for applications is 28 February, 2020.

At the UK general election on 12 December the exit poll, which is commissioned jointly by broadcasters BBC, ITV and Sky, will again use the methods that were introduced in 2001 by Professor David Firth (in joint work with Professor John Curtice of the University of Strathclyde).

The new methods have led to renewed trust in exit polls, which since 2001 have been remarkably accurate in predicting House of Commons seat totals — a prediction made at 10pm on election day, immediately after polling stations close their doors.

David Firth says:

It's great to see the approach that I suggested and developed in 2001–2005 being used still in 2019. Its success is due mainly to purposeful statistical thinking in both the design and the analysis of the exit poll; and partly of course also to luck! But it could not succeed without, in addition, the work of a brilliant team of academics on the election day itself.

The election-day team analysing the 2019 exit poll will again be led by Sir John Curtice. The team's statistical expert has since 2010 been Professor Jouni Kuha (LSE) — a good friend of David Firth's since the 1990s. David says of this: "Perhaps the most remarkable thing, for me personally, is that Jouni can still manage to run the same R code that I threw together in haste before the 2005 election!"

For more information on the exit-poll method and its track record, see the Exit Poll Explainer page at warwick.ac.uk/exitpolling.