Welcome to the News and Events page for the Department of Statistics.
Election exit poll: Not quite 'spot on' this time, but another triumph for statistical methods!
The exit-poll design and analysis methods developed by David Firth (with political scientist John Curtice from the University of Strathclyde) were used again at this week's General Election by all of the major UK broadcasters.
At 10pm on election day the on-air seats prediction (simultaneously on BBC, ITV and Sky) based on the exit poll was: Con 316, Lab 239, SNP 58, LD 10, others 27. The actual result of the election was Con 331, Lab 232, SNP 56, LD 8, others 23.
The 2015 exit-poll prediction was thus not "spot on" as it had been in 2005 and 2010. Many commentators had warned beforehand that the 2015 election would be an especially difficult one to predict. The exit-poll prediction was startlingly different from what had been indicated by commercial pre-election voting-intention polls. (e.g., see The Observer on 10 May, After the exit poll, a tsunami raged across the political map) The exit poll strongly indicated the Conservatives as largest party, and the ultimate outcome of a small Conservative majority was clearly not ruled out. This was in stark contrast to predictions from pre-election polls, which had consistently shown Conservative and Labour neck-and-neck with neither party close to an overall majority.
There were some notable public quotes, most prominent of which came from the former Liberal Democrat leader Lord Ashdown, who was interviewed soon after 10pm on the BBC:
If this exit poll is right, I will publicly eat my hat on your programme.
(He was referring to the predicted collapse of the Liberal Democrats to just 10 parliamentary seats. In the event, it turned out even worse than that for the Liberal Democrats, who won just 8 seats. Lord Ashdown failed to keep his hat-eating promise, though!)
For more information on the methods and their performance at previous UK general elections, see Exit Polling Explained.
Warwick selected as one of the founding members of the Alan Turing Institute for Data Science
The Department of Statistics at Warwick led a team of academics from the Mathematics Institute and the Departments of Computer Science and Statistics in bidding to be founding partners in establishing the Alan Turing Institute for Data Science in the UK. The Business Secretary, Vince Cable, has announced that five universities were selected to lead the Alan Turing Institute, one of which is Warwick. The four other universities are Oxford, Cambridge, UCL and Edinburgh. Warwick’s pivotal role in the Alan Turing Institute will be led by Professor Mark Girolami from the Department of Statistics.
Data Science has been an integral part of Warwick’s vision for the Mathematical Sciences. Warwick is home to the Warwick Data Science Institute and an undergraduate degree programme in Data Science, both the first of their kind in the UK.
The Alan Turing Institute will be physically located in London’s new Knowledge Quarter.
Heather Turner to give invited course in Zurich
Dr Heather Turner, statistical consultant and currently an Associate Fellow of Warwick Statistics, has been invited to give an intensive one-day course Introduction to Generalized Nonlinear Models in Zurich on 11 May 2015. The course is based on Heather's work as a Senior Research Fellow at Warwick, which included the development of award-winning software package gnm.
Heather's work and her views on statistics, science and R development are featured in this extended interview recorded at the useR! 2014 conference in Los Angeles.
IMS Blackwell Lecture award to Professor Gareth Roberts
At the Joint Statistical Meetings in Boston in early August, Gareth Roberts will give the inaugural Blackwell Lecture, awarded by the Institute of Mathematical Statistics. The lecture has been introduced in honour of the extraordinary career in statistics and probability of Professor David Blackwell, who died in 2010 at the age of 91. The lecture should present material with some connection to David Blackwell's work, although since Blackwell's work was so broad, this turns out not to be very restrictive. Gareth's lecture will talk about work on Monte Carlo methods for stochastic processes, which has close connections to one of Blackwell's most famous theorems: The Rao-Blackwell Theorem.
It is a tremendous honour for Gareth to be chosen as the first-ever Blackwell Lecturer. The title and abstract of his lecture are given below.
Rao-Blackwellization for Improved Monte Carlo for Stochastic Processes
David Blackwell's celebrated theorem with Rao compares the variance of an estimator X with that of its conditional expectation given a further variable S; that is, Y = E(X|S). Blackwell's motivation was to demonstrate that the best estimators should be functions of the simplest sufficient statistic we can find. However the Rao-Blackwell theorem has also found extensive use in computational statistics by providing a powerful and general recipe for improving on Monte-Carlo estimators by so-called Rao-Blackwellisation. This talk will discuss widely adopted, simple Rao-Blackwellisations for Markov chain Monte Carlo and Sequential Monte Carlo samplers, as well as more complex random Rao-Blackwellisations for estimation of expectations of functionals of stochastic processes (particularly univariate and multivariate diffusion processes).
The lecture will draw on joint research with Alex Beskos, Paul Fearnhead, Krys Latuszynski, Omiros Papaspiliopoulos and Giorgos Sermaidis.
Warwick Statistics has two professors in the 2014 worldwide "Highly Cited" list
Gareth is listed as one of just 100 top researchers worldwide in the "Mathematics" category, and Tom appears as one of just 129 Highly Cited scientists in the "Neuroscience and Behavior" category.