Abstract TT2011
Trinity Term 2011
Week 2: 12 May
Speakers: Alan Smith and Steven Rogers (Data Visualisation Centre, Office for National Statistics)
Title: ONS Data Visualisation Centre: From Data to Meaningful Information (slides)
Abstract: Statistical information has long played an important role in shaping public social and economic policy. But increasing calls for more transparency (‘open data’) and wider access to meaningful information pose a significant challenge to producers of official statistics. While the internet offers an important gateway for wider access to data, the dissemination of even basic statistical information is not a trivial task, particularly when the audience has little or no formal statistical training.
“Numbers are everywhere. But mostly we don't really get what they mean, even when they're key to the important choices we make in our lives.” (RSS getstats campaign, 2011)
Drawing on long established disciplines such as Visual Perception, Cartography, Human Computer Interaction, and Colourimetry, this session will look at some of the reasons why information encoded in a visual interface such as a graph or map may not be understood accurately or efficiently. We also demonstrate how ONS is developing presentation standards, techniques and tools that improve the way information in both static and interactive statistical graphics is perceived.
Week 3: 19 May
Speaker: Andrew Dalby, Bioinformatician at BBSRC Research Institute
Title: Are spatial stochastic models relevant for modelling cellular processes?
Abstract: Systems biology has focused on trying to understand biology at the cellular level. From molecular biology we have a good knowledge of the building blocks of biochemical systems but we need to build integrated models that can explain cellular level phenomena such as signalling or metabolic control. Experimental data has shown that cells exhibit stochastic variation in properties such as gene expression and chemotactic movements. Stochastic models can be built based on the chemical master equation, but these do not take into account spatial effects, that might be significant in biology. So a general question is whether spatial stochastic models can provide improved insight into cellular processes.
Week 4: 26 May
Speaker: Venkat Chandrasekaran, MIT
Title: Rank/Sparsity Minimization and Latent Variable Graphical Model Selection?
Abstract: Suppose we have a Gaussian graphical model with sample observations of only a subset of the variables. Can we separate the extra correlations induced due to marginalization over the unobserved, hidden variables from the structure among the observed variables? In other words is it still possible to consistently perform model selection despite the unobserved, latent variables?
As we shall see the key problem that arises is one of decomposing the concentration matrix of the observed variables into a sparse matrix (representing graphical model structure among the observed variables) and a low rank matrix (representing the effects of marginalization over the hidden variables). Such a decomposition can be accomplished by an estimator that is given by a tractable convex program. This estimator performs consistent model selection in the high-dimensional scaling regime in which the number of observed/hidden variables grows with the number of samples of the observed variables. The geometric aspects of our approach are highlighted, with the algebraic varieties of sparse matrices and of low-rank matrices playing an important role.
Joint work with Pablo Parrilo and Alan Willsky (MIT).
Week 5: 2 June
Speaker: Chris Jewell, Statistics Department, Warwick University
Title: Bayesian risk prediction for livestock epidemics
Abstract: The unpredictable nature of infectious disease epidemics implies that the characteristics of an outbreak are hard to predict in advance of an incursion. Changes in the host, pathogen, and environment over time, may mean that a new outbreak may not behave as expected based on prior information alone. Therefore, predictions based on current epidemic field data are required for tailoring control-policy to the current outbreak.
In trying to gain a more accurate insight into how an outbreak might spread through a population, mathematical simulation modelling has become a popular and established tool. Yet simulations rely on knowing certain parameters governing epidemic dynamics, and estimating appropriate values from outbreak data has been a major challenge to their predictive credibility. Addressing this, this talk will describe a generic rjMCMC based Bayesian framework for analysing epidemic data in real time, estimating critical disease parameters, and imputing individuals' necessarily unobserved infection times. I will present examples on Avian Influenza in Poultry and foot and mouth disease in cattle to illustrate how this framework can be applied to different disease outbreak scenarios, and how Bayesian inference can be used in conjunction with simulation and GIS techniques to provide information for decision-making during epidemics.
Week 7: 16 June
First year DPhil poster session in 2 South Parks Road
Printer Friendly Version 