Centre for Statistics

Members

Our core members are data-driven researchers from across Edinburgh, who are involved in the promotion and application of the statistical sciences.

The Centre for Statistics is led and overseen by the Executive Committee and the Steering Group. View the organisation structure.

To become a member, please contact DirectorCfS[at]ed.ac.uk

For more information, or to join our mailing lists, please contact us here CfS-comms[at]ed.ac.uk

Quick Jump:  A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

photograph of Colin Aitken

Colin Aitken

Research interests

The interface of law, statistics and forensic science.

Keywords (methods)

Multivariate hierarchical Bayesian models

Keywords (applications)

Evidence evaluation and interpretation, forensic statistics

photograph of Mike Allerhand

Mike Allerhand

Research interests

Statistical modelling, machine learning, statistical programming, and numerical analysis. Mike is an expert R programmer and has skills in data visualization, data reduction, feature engineering, statistical modelling, and machine learning for inference, prediction, and classification.

Keywords (methods)

generalised and additive regression models, Bayesian inference, networks and graphs, state-space methods, machine learning, pattern recognition

Keywords (applications)

speech and auditory modelling, cognitive ability and personality, biostatistics, neuropsychology

photograph of Niall Anderson

Niall Anderson

Research interests

Medical statistician, working in collaborative research within public health and clinical medicine. Range of experience across statistical genetics and genetic epidemiology, inflammatory bowel disease research, prediction of ventilator-acquired pneumonia, as well as more general clinical research topics. Heavily involved in statistical education and consulting.

Keywords (methods)

modelling, diagnostic tests, mixed models, teaching, consultancy

Keywords (applications)

public health, epidemiology, clinical trials, consulting, education

photograph of Galina Andreeva

Galina Andreeva

Research interests

My research evolves around credit risk of individuals and small businesses (SMEs) using statistical and machine-learning techniques, applied to Big Data, in particular Open Banking (OB) financial transactions. An area of specific focus is financial vulnerability, fairness and social impact of credit.

Keywords (methods)

Keywords (applications)

photograph of Jake Ansell

Jake Ansell (Chair, Steering Committee)

Research interests

Applied statistician/management scientist applying statistical and optimisation methodologies to problems in the domain of business and engineering. Aim is to develop new approaches to issues that relate to enhancing performance of systems and exploring issues across range of areas.

Keywords (methods)

reliability and maintenance, credit scoring, risk management, optimisation

Keywords (applications)

banking, financial capability, engineering (reliability), SMEs, marketing

photograph of Nicole Augustin

Nicole Augustin

Research interests

My research is mainly in the areas of spatio-temporal modelling of natural resources (forest health; fisheries), health statistics (intensively collected health data; tobacco control) and model selection uncertainty. Recent work is on spatial confounding in the context of modelling environmental data.

Keywords (methods)

spatio-temporal models, spatial models, Health statistics, Spatial confounding, Model selection uncertainty

Keywords (applications)

ecology, Accelerometer data, Fisheries stock management, Forest health monitoring, Physical activity, Tobacco control

photograph of Daniel Barker

Daniel Barker

Research interests

Bioinformatics education, phylogeny, comparative genomics, philosophy of science

Keywords (methods)

Likelihood

Keywords (applications)

comparative genomics, evolutionary biology, functional genomics

photograph of Hannes Becher

Hannes Becher

Research interests

Statistical genetics and genomics

Keywords (methods)

Mixed-effect models

Keywords (applications)

demography, Natural selection, adaptation

photograph of Sjoerd Beentjes

Sjoerd Beentjes

Research interests

I am interested in applications of pure mathematics and mathematical statistics to causal questions in population biomedicine and public health policy. In particular, I am interested in developing mathematical and statistical techniques in the framework of Targeted Learning to extract precise answers to causal biological questions from large population-scale databases, such as the UK Biobank and Generation Scotland.

Keywords (methods)

causal inference, model-independent and nonparametric methods, Targeted Learning, topological data analysis

Keywords (applications)

population genetics, public health policy, uncovering causal variants leading to complex disease or trait

photograph of Michele Belot

Michele Belot

Research interests

Evaluation of economic policies in the domains of health, labour and education. Currently my two main research agendas relate to dietary habits and advice to job seekers and in both cases I am interested in studying behavioural change.

Keywords (methods)

difference-in-differences analysis, discrete choice models, experiments, survival analysis

Keywords (applications)

dietary choices, educational achievements, job search behaviour

photograph of Natalia Bochkina

Natalia Bochkina

Research interests

My research interests include statistical theory, methodology and applications. I am interested in developing statistical methods for challenging applied problems, and making them efficient and robust. I am also interested in developing robust statistical inference based on fast approximate methods, e.g. coming from machine learning

Keywords (methods)

Bayesian inference, nonparametric inference, wavelet methods, inverse problems, model misspecification, graphical models

Keywords (applications)

biology, genomics, medicine, geoscience

photograph of Tom Booth

Tom Booth

Research interests

I am an applied quantitative methods researcher with a primary interest in individual differences, organisational psychology and health. I am interested in the application of statistical models (broadly speaking generalized latent variable models) to large scale, secondary, longitudinal data.

Keywords (methods)

structural equation modelling, psychometrics & generalized latent variable models, longitudinal data

Keywords (applications)

epidemiology, individual differences, organisational psychology, health

photograph of Mark Brewer

Mark Brewer (Steering Committee member)

Research interests

Working with ecological and environmental scientists to solve real-world problems. Applying Bayesian and non-Bayesian models as appropriate, and with recent concern over issues of model selection and comparison - especially for between-data set heterogeneity. Other recent work looks at constrained optimisation for ordinal response models.

Keywords (methods)

model assessment and comparison, Bayesian spatio-temporal modelling, ordinal response models, compositional modelling

Keywords (applications)

ecology, food authenticity, water quality, soil forensics, species distribution, socio-economics

photograph of Andrew Cairns

Andrew Cairns

Research interests

Research into the modelling, measurement and management of mortality and longevity risk in life insurance and pensions. Development of the best models for understanding historical mortality improvements, and for predicting future mortality rates. Identification of mortality inequalities across national populations.

Keywords (methods)

Bayesian methods, financial mathematics, simulation, stochastic modelling

Keywords (applications)

cause of death analysis, financial risk management, insurance, model risk, mortality and longevity risk, mortality inequality, pensions, predictive analysis for mortality

photograph of Tim Cannings

Tim Cannings (Director)

Research interests

My research interests are broadly in the area of statistical methodology and theory, including statistical learning problems, nonparametric methods, high-dimensional data and semi-supervised learning. My work is primarily motivated by the recent, rapid, developments in technology, which have resulted in new complex data structures. In particular, we now have the ability to collect and store vast amounts of data routinely, this poses new challenges for statisticians and data scientists.

Keywords (methods)

high-dimensional, nonparametric, random projections, data perturbation, noisy and incomplete data

Keywords (applications)

big data analytics, cancer therapy, genomic data

photograph of Mine Çetinkaya-Rundel

Mine Çetinkaya-Rundel

Research interests

Statistical computing, statistics and data science education

Keywords (methods)

Keywords (applications)

photograph of Zexun Chen

Zexun Chen

Research interests

My research interests focus on probabilistic machine modelling and non-parametric Bayesian predictive methods such as Gaussian Process and Student-t Process modelling. As their applications, I also have expertise in time series prediction, financial data analysis, algorithmic fairness, and human mobility.

Keywords (methods)

Bayesian Non-Parametric Modelling, Gaussian process, time series modelling

Keywords (applications)

algorithmic fairness, dam structure monitoring, Financial market risk, human mobility, social network analysis

photograph of Alfred Chong

Alfred Chong

Research interests

Actuarial Science, Financial Mathematics, Quantitative Risk Management

Keywords (methods)

Reinforcement Learning, Supervised Learning

Keywords (applications)

Cyber Risk Management, Life and Retirement Products, Risk Aggregation and Resources Allocation, Risk Sharing

photograph of Damian Clancy

Damian Clancy

Research interests

Stochastic modelling of infectious disease spread, including epidemic models and models for endemic infections; the effects of population heterogeneities on infectious spread; veterinary applications. Bayesian inference, particularly for infectious disease models, and also for ecological models.

Keywords (methods)

Bayesian inference, stochastic models, Markov processes, epidemic models

Keywords (applications)

epidemiology, veterinary epidemiology, ecology

photograph of Nick Colegrave

Nick Colegrave

Research interests

My research interests are in the interface between ecology and evolution. I am interested in developing simulation approaches to power analysis and improving experimental design and statistics among biological researchers.

Keywords (methods)

Power analysis by simulation

Keywords (applications)

ecology, evolution, statistics education

photograph of Helen Colhoun

Helen Colhoun

Research interests

Using electronic health data to model the risk of diabetes complications. Availing of linkage to biobanks to generate and examine other high dimensional data including molecular and 'omics data for prediction. Focus on understanding the pathogenesis and means of prevention of diabetic complications with studies ranging from genetic epidemiology and biomarker studies through to clinical trials. Large scale epidemiological analyses for informing health policy.

Keywords (methods)

survival analyses, growth curve analyses, prediction performance, genome wide association studies (GWAS)

Keywords (applications)

epidemiology, genetics, diabetes, cardiovascular disease, kidney disease

photograph of Roxanne Connelly

Roxanne Connelly

Research interests

Social Stratification, Sociology of Education, Social Inequalities, Analysis of Large and Complex Social Science Data Resources

Keywords (methods)

longitudinal data analysis, Complex Samples, Social Survey Data Analysis, Latent Class Analysis

Keywords (applications)

Social Stratification, Social Inequalities

photograph of Jonathan Crook

Jonathan Crook

Research interests

My interests are in exploring statistical and econometric issues in all aspects of credit risk. These issues include sample selection problems, model risk, dynamic models, missing data issues and involve parametric and non-parametric methodologies . The application areas typically involve probability of default, loss given default, exposure at default and stress testing.

Keywords (methods)

APC models, Bayesian inference, classification, intensity models, machine learning, missing data, mixed models, panel models, sample selection models, spatial models, survival analysis

Keywords (applications)

bank capital modelling, credit scoring, stress testing

photograph of Andrew Curtis

Andrew Curtis

Research interests

I focus mainly on Bayesian design, image creation, and image interpretation problems. I also study how to obtain robust a priori information from humans to drive Bayesian inference and decision making processes. I tend to create images of the structure or properties of media using acoustic or elastic waves, or electrical resistivity data. I develop optimal elicitation methods and interrogation theory to analyse uncertainties in human-derived information and decisions.

Keywords (methods)

Bayesian inference, optimisation, expert elicitation, design, interrogation, Monte Carlo, hidden Markov models, wave physics

Keywords (applications)

geoscience, geophysics, geology, non-destructive testing, medical imaging, seismic, acoustic, ultrasound, resistivity

photograph of Jean-François Daoust

Jean-François Daoust

Research interests

I am a social scientist who make use or statistics to study public opinion (satisfaction with democracy, etc.), elections, voting behavior (turnout, vote choice). I am particularly interested in experimental and quasi-experimental research designs.

Keywords (methods)

statistical modelling, experimental design, regression discontinuity design, survey data

Keywords (applications)

Social sciences; politic; public opinion; elections; communication

photograph of Miguel de Carvalho

Miguel de Carvalho

Research interests

My research interests are eclectic. I am an applied mathematical statistician with a variety of research interests including, inter alia, statistical inferences for small-probability events, geometrical statistics, methods for data visualization and graphical learning, econometrics, and medical diagnostic assessment.

Keywords (methods)

Bayesian inference, nonparametric inference, empirical likelihood, singular spectrum analysis, statistics of extremes

Keywords (applications)

medicine, business, macroeconomics, finance, risk

photograph of Chris Dent

Chris Dent (Co-Director, Consultancy)

Research interests

Energy systems analysis. Electricity security of supply risk. Renewable resource modelling. Use of modelling and data in government, including model calibration, uncertainty analysis, role of modelling within decision processes, and communication of results beyond the technical modelling community. Industrial applications of mathematics.

Keywords (methods)

statistical emulation, uncertainty quantification, time series, optimization, applied probability

Keywords (applications)

risk, energy, electricity, government, communication, renewables, variable generation, model calibration, industrial applications

photograph of Goncalo dos Reis

Goncalo dos Reis

Research interests

My research is at the interface of applied probability, stochastics, and statistics. Currently I am investigating applications of machine learning to a variety of problems in financial risk and have several projects with financial industry partners on topics of model calibration, risk computation, model development and software development.

Keywords (methods)

Hawkes processes and point processes, hidden Markov models, high-dimensional sampling algorithms, large deviations principles, Markov chain Monte Carlo and simulation, mean-field and related techniques in Wasserstein spaces, Monte Carlo, optimal control, transportation and concentration of measure, uncertainty quantification

Keywords (applications)

banking risk, game theory, machine learning, mathematical finance, probabilistic numerics, risk analysis, risk management

photograph of Dimitrios Doudesis

Dimitrios Doudesis

Research interests

Dimitrios is a health data scientist with a background in statistics. He is an R expert with a primary interest in using electronic health data to model the risk of acute cardiac conditions. He is working with clinicians to solve real-world problems using statistical machine learning.

Keywords (methods)

machine learning, data science, statistical theory, diagnostic and prognostic modelling, feature selection

Keywords (applications)

biostatistics, medicine, precision medicine, electronic health records, medical statistics, disease risk stratification, diagnostic tests, biomarker

photograph of Meriem El Karoui

Meriem El Karoui

Research interests

My research focuses on the integration of bacterial physiology in the understanding of the molecular processes underlying DNA maintenance and antibiotic tolerance. I have established a group of researchers with merged expertise from mathematics, microbiology and biophysics.

Keywords (methods)

stochastic modelling

Keywords (applications)

DNA repair, genetics, microbiology

photograph of Victor Elvira

Victor Elvira

Research interests

My research interests are mostly at the intersection of computational statistics and statistical signal processing, in particular in Monte Carlo methods for Bayesian inference (e.g., importance sampling, Markov chain Monte Carlo, particle filtering). I have also developed a recent interest in statistical methods that incorporate optimization tools for a faster learning.

Keywords (methods)

Bayesian inference, Monte Carlo methods, state-space models, computational statistics, statistical statistics

Keywords (applications)

remote sensing, tracking, localization, wireless sensor networks, communications

photograph of Ailith Ewing

Ailith Ewing

Research interests

We use statistical genomics to investigate the evolution of large scale rearrangements in the DNA of tumours to inform patient stratification, better predict treatment response and improve clinical outcome. Interrogating complex tumour genomes involves the analysis of complex high-dimensional datasets and relies on the effective application of methods to reduce the dimensionality of the data and extract meaningful signal from the extensive heterogeneity.

Keywords (methods)

clustering, Dimension reduction, high-dimensional regression

Keywords (applications)

genomics, genetics, cancer, biomedicine

photograph of Kelly Fleetwood

Kelly Fleetwood

Research interests

I am interested in the relationship between physical and mental health. I use data from cohort studies and routine health data (e.g. hospital records) to explore the associations between severe mental illness and the incidence and outcomes of physical illnesses. I previously worked on meta-analysis and network meta-analysis of data from randomized clinical trials.

Keywords (methods)

statistical modelling, meta-analysis, network meta-analysis

Keywords (applications)

statistical modelling, meta-analysis, network meta-analysis

photograph of Jonathan Gair

Jonathan Gair

Research interests

Development of data analysis techniques for and scientific exploitation of observations made by gravitational wave detectors, including LIGO, LISA, pulsar timing and the Einstein Telescope. Statistical emulation of complex process models in physics and environmental science. Assessment of ecosystem services provided by habitat restoration

Keywords (methods)

Bayesian inference, statistical emulation, uncertainty quantification, Gaussian processes, hierarchical models

Keywords (applications)

ecology, astrophysics, gravitational waves, LIGO, LISA, pulsar timing, hydrology, NIRAMS

photograph of Vernon Gayle

Vernon Gayle

Research interests

I work with large-scale and complex social science datasets. I am especially interested in the emerging field of social data science.

Keywords (methods)

longitudinal data analysis, statistical modelling

Keywords (applications)

education, sociology, data science

photograph of Gavin Gibson

Gavin Gibson

Research interests

My research deals with methods of parameter estimation and model assessment for spatio-temporal stochastic models with applications mainly in infectious disease epidemiology where particular challenges arise due to the typically incomplete nature of observations. Working mainly in the Bayesian framework, while drawing on classical methods, I make extensive of data augmentation and computational methods. My work has involved extensive collaboration with biological and medical scientists.

Keywords (methods)

Bayesian inference, Markov chain Monte Carlo, model assessment and comparison, posterior predictive methods, spatio-temporal models

Keywords (applications)

arboreal pathogens, built environment, image processing, laser imaging systems, phylodynamics, veterinary pathogens

photograph of Gregor Gorjanc

Gregor Gorjanc

Research interests

Development and application of statistical and computational methods for use in conjunction with genetics and breeding to manage and improve populations. I am specifically interested in: (i) applied breeding, (ii) design and optimisation of breeding programs, (iii) methods for population and quantitative genetics and breeding, and (iv) analysis of complex traits to unravel their biological basis and to inform new ways of breeding.

Keywords (methods)

applied Bayesian statistics, high performance computing, methods for high-dimensional covariate and observation data, mixed/hierarchical/latent models

Keywords (applications)

agriculture, biotechnology, breeding, population, statistical and quantitative genetics/genomics

photograph of Michael Gutmann

Michael Gutmann

Research interests

Developing methods for efficient statistical inference and model-based experimental design and using them to solve challenging problems in the natural sciences.

Keywords (methods)

approximate Bayesian computation, likelihood-free inference, Bayesian experimental design, implicit models, unnormalised/energy-based models, unsupervised machine learning

Keywords (applications)

computational biology, epidemiology of infectious diseases

photograph of Jarrod Hadfield

Jarrod Hadfield

Research interests

I am an evolutionary biologist mainly working in the area of quantitative genetics. I develop statistical methods and algorithms for analysing data commonly collected by evolutionary biologists, some of which are general (for example, hierarchical models for non-Gaussian data) and some of which are more specific (for example, pedigree reconstruction from genetic markers).

Keywords (methods)

Bayesian inference, hierarchical models, pedigree reconstruction, phylogenetic comparisons

Keywords (applications)

ecology, quantitative genetics, evolutionary biology

photograph of Ewen Harrison

Ewen Harrison

Research interests

I am Director of the Centre for Medical Informatics within the Usher Institute, University of Edinburgh and a Consultant Surgeon. My data-driven research focuses on understanding disease and improving patient outcomes after surgery.

Keywords (methods)

Bayesian methods (Stan), causal inference, crowd-sourcing patient-level data, decision/prognostic modelling, machine learning, mobile data collection platforms and “wearables”, patient-reported outcomes, surgical trial methodology

Keywords (applications)

epidemiology, medicine

photograph of Matthew Hartfield

Matthew Hartfield

Research interests

I am an evolutionary geneticist who uses mathematical and computational methods to infer how fundamental evolutionary forces (selection, reproduction and so on) have acted from genome sequence data. More broadly I am also interested in data science and applying such methods to broader biological databases as part of my research.

Keywords (methods)

hidden Markov models, Likelihood, machine learning, model selection

Keywords (applications)

Evolutionary biology and genomics, inferences of selection, mating system

photograph of Fengxiang He

Fengxiang He

Research interests

My research interest is in trustworthy AI, particularly deep learning theory and explainability, theory of decentralised learning, privacy in machine learning, symmetry in machine learning, learning theory in game-theoretical problems, and their applications in economics.

Keywords (methods)

algorithmic fairness, algorithmic game theory, differential privacy, learning theory, machine learning, mechanism design, neural network

Keywords (applications)

computer vision, economics, finance, health care

photograph of Gabriele Hegerl

Gabriele Hegerl

Research interests

Detecting, attributing and understanding observed climate variability and change. Focus on variability and change in climatic extremes and precipitation, estimating climate sensitivity, and use of palaeo proxy data to study climate variability and change during the last millennium

Keywords (methods)

attribution methods, Bayesian analysis, orthogonal functions, Monte Carlo methods

Keywords (applications)

statistical climatology, climate change, extreme weather and climate events

photograph of Vanda Inacio de Carvalho

Vanda Inacio de Carvalho

Research interests

Development of (Bayesian) flexible methods and software for biostatistical applications, in particular, for the evaluation of medical tests/biomarkers.

Keywords (methods)

Bayesian inference, computational statistics, nonparametric methods

Keywords (applications)

biostatistics, medicine

photograph of Tony Kennedy

Tony Kennedy

Research interests

I am a theoretical physicist working mainly in the area of quantum field theory, with a particular interest in Markov Chain Monte Carlo algorithms such as Hamiltonian Monte Carlo (HMC). I have recently been working on developing practical methods for HMC on non-trivial manifolds such as symmetric spaces. It appears that quantum field theory has surprising connections with stochastic differential equations, and I am interested in understanding these connections better and perhaps applying techniques, such as renormalization, developed in one field to the other.

Keywords (methods)

Hamiltonian Monte Carlo, Markov chain Monte Carlo, uncertainty quantification

Keywords (applications)

quantum field theory, theoretical physics

photograph of Ava Khamseh

Ava Khamseh

Research interests

I'm a Lecturer in Biomedical Artificial Intelligence, joint between the School of Informatics and Institute of Genetics and Cancer (IGC). Before taking up this position, I was a Cross-Disciplinary Fellow (XDF) at IGC. My research involves designing experiments and developing causal mathematical, statistical and machine learning methods for applications to cancer biology and large scale population biology data.

Keywords (methods)

Targeted Learning, Causal machine learning, model-independent estimation, model-misspecification

Keywords (applications)

population genetics, Cancer initiation and evolution

photograph of Ruth King

Ruth King

Research interests

I am an applied statistician, developing new statistical methodology motivated by real problems, primarily within the areas of ecology and healthcare. Particular interests include constructing novel statistical models, involving for example individual heterogeneity, memory and integrated data analyses; and associated efficient model-fitting algorithms

Keywords (methods)

Bayesian inference, capture-recapture, hidden (semi-)Markov models, integrated models, missing data, state-space models

Keywords (applications)

epidemiology, medicine, ecology

photograph of Stuart King

Stuart King

Research interests

I work in applied mathematics with an interest in data science techniques and problems. In particular I am interested in developing statistical and machine learning techniques to images and image sequences applied to data in a range of areas, including satellite and aircraft based Earth observation through to medical imaging.

Keywords (methods)

clustering, classification, change detection, image processing

Keywords (applications)

ecology, remote sensing, satellite images, medical applications

photograph of Aleksander Kolev

Aleksander Kolev

Research interests

N/A

Keywords (methods)

State-Space models; Time series; Hawkes process; ETAS model; Bayesian inference; Capture-recapture; Clustering;

Keywords (applications)

photograph of Alison Koslowski

Alison Koslowski

Research interests

I generally work with large scale cross-national social survey data in the pursuit of a better understanding of questions concerning the labour market, care work, and gender equality.

Keywords (methods)

social science survey data (data collection and analysis), longitudinal data analysis

Keywords (applications)

social policy, sociology, demography

photograph of Amanda Lenzi

Amanda Lenzi

Research interests

My main research interests concern statistical modeling, prediction, simulation, and uncertainty quantification of spatiotemporal data. I also work on developing computational methods for large datasets and the use of deep learning to improve the inference of models with complex dependencies.

Keywords (methods)

Bayesian inference, deep learning, spatiotemporal statistics, statistics of extremes

Keywords (applications)

energy, environmental science, geoscience

photograph of Steff Lewis

Steff Lewis

Research interests

I am one of the leads of the statistics team in Edinburgh Clinical Trials Unit, and my research interests are around many practical areas of design, conduct and analysis of randomised trials. I've been particularly involved in guidance for Statistical Analysis Plans and anonymisation of trial datasets, and choice of minimisation variables (and the use of random elements in minimisation). I also dabble in meta-analysis and prognostic models.

Keywords (methods)

anonymisation, baseline adjustment, meta-analysis, randomisation

Keywords (applications)

clinical trials, medicine

photograph of You Li

You Li

Research interests

I am an epidemiologist, with broad interest in infectious diseases epidemiology and modelling, global health and child health. I am an R programmer with skills in statistical modelling, machine learning, data imputation and meta-analysis. I develop R packages and R ShinyApps.

Keywords (methods)

meta-analysis, Health statistics, Prediction model, infectious disease modelling, data imputation

Keywords (applications)

Disease burden estimate, infectious disease epidemiology and prediction, global health, child health

photograph of Finn Lindgren

Finn Lindgren

Research interests

Developing probabilistic models for spatial and space-time phenomena, suitable for ecological, climatological, and general environmetrics applications, as well as medical imaging problems. Computational solutions for Bayesian statistics with hierarchical generative models.

Keywords (methods)

Bayesian inference, Gaussian processes, random fields, Gaussian Markov random fields, stochastic partial differential equations, fast approximate inference

Keywords (applications)

ecology, medical imaging, animal abundance, biodiversity, climatology, historical temperature reconstruction, fMRI

photograph of Samantha Lycett

Samantha Lycett

Research interests

Evolution and epidemiology of viruses and bacteria, including machine learning techniques and Bayesian phylogenetics to investigate cross species transmissions, host adaptations, epistatic interactions, phylodynamics and phylogeography. I'm developing fast computational methods and simulation tools to infer transmission patterns of animal and livestock pathogens (including in missing data situations), and methods to extract predictive factors for disease risk and new strain generation

Keywords (methods)

Bayesian inference, phylodynamics, phylogeography, machine learning methods, network inference, approximate methods, stochastic simulations

Keywords (applications)

epidemiology, sequence data, viruses, bacteria, animal and zoonotic pathogens, disease networks, pathogen evolution

photograph of Ian Main

Ian Main

Research interests

Processes that lead up to catastrophic failure events, from earthquakes, rock fracture, and volcanic eruptions to failure of building materials and bridges. Population dynamics of localised brittle failure as a complex, non-linear (unreasonable) system, and in quantifying the resulting hazard

Keywords (methods)

Bayesian analysis, Monte Carlo methods, multivariate statistics, epidemic-type stochastic models for point processes

Keywords (applications)

probabilistic seismic hazard estimation, prospective earthquake forecasting, communicating risk and uncertainty, hazard mitigation and disaster risk reduction.

photograph of Nicolo Margaritella

Nicolo Margaritella

Research interests

My research focuses on the development of new statistical methodology motivated by real problems, primarily within the areas of neuroscience and healthcare. I am particularly interested in the analysis of complex spatio-temporal data and large-scale inference problems.

Keywords (methods)

Bayesian inference, functional data analysis, large-scale inference, spatio-temporal models

Keywords (applications)

medicine, neuroscience

photograph of Riccardo Marioni

Riccardo Marioni

Research interests

Ageing, health, brain health, omics (genetics, epigenetics, proteomics), data linkage, longitudinal data

Keywords (methods)

linear, logistic, mixed models, penalised, Regression, survival

Keywords (applications)

ageing, brain health, data linkage, epigenetics, genetics, health, longitudinal data, omics, proteomics

photograph of Alan Marshall

Alan Marshall

Research interests

My substantive research uses longitudinal data from social surveys in the UK and overseas to better understand the social and biological determinants of inequalities observed in health and wellbeing in later life. My methodological research contributions centre around the development of local estimates and projections of populations and of populations in poor health in collaboration with the UK's National Statistical Agencies and Local Authorities.

Keywords (methods)

social science survey data (data collection and analysis), longitudinal data analysis, census data, data linkage, applied demography

Keywords (applications)

social policy, sociology, demography, inequality, international comparisons, ageing

photograph of Paul McKeigue

Paul McKeigue

Research interests

Methods for molecular and genetic epidemiology, with applications in clinical prediction and personalized medicine. These include using biomarkers or automated scoring of images to predict diabetic complications, to predict drug response in rheumatoid arthritis, and to detect cancer in national screening programmes.

Keywords (methods)

Bayesian hypothesis testing, Bayesian computation, Markov chain Monte Carlo, variational inference, hierarchical shrinkage priors, projection predictive variable selection, deep learning

Keywords (applications)

biomarker discovery, biomarker selection, genetic prediction, precision medicine, risk stratification, image classification

photograph of Susan McVie

Susan McVie

Research interests

My main area of research currently is around crime and justice inequalities which includes using large survey and administrative datasets to explore issues around the impact of poverty, deprivation and unequal access to services on criminal behaviour and patterns of victimisation. I am also involved in developing new areas of data linkage between law enforcement and public health organisations. I manage a large longitudinal study of offending behaviour and I am involved in developing new areas of research around violent criminal careers

Keywords (methods)

structural equation modelling, longitudinal methods and panel analysis, multi-level analysis, quasi-experimental methods

Keywords (applications)

international crime trends and patterns, youth crime and juvenile justice, criminal careers through the life-course, patterns of violence and homicide, youth gangs and knife crime, policing and crime reduction, stop and search, police use of biometric data

photograph of Franziska Meinck

Franziska Meinck

Research interests

Survey research, working with triadic/dyadic data, casual inference, measurement development and validation

Keywords (methods)

SEM, quasi-experimental designs, surveys

Keywords (applications)

public health, epidemiology, psychometrics

photograph of Ben Moews

Ben Moews

Research interests

My research is centred on the intersection of machine learning, statistical inference and high-performance computing with domain applications. Past and current work includes high-dimensional parameter estimation, machine and deep learning for simulation emulators and generative models, applications of spatio-temporal statistics and principal curves, and statistical software.

Keywords (methods)

Bayesian inference, computational statistics, machine learning, sampling methods, spatial analysis

Keywords (applications)

astronomy, criminology, finance, healthcare

photograph of Iain Murray

Iain Murray

Research interests

I develop computational methods to perform probabilistic reasoning about unknown parameters, or actions in unknown or changing environments. As these models can be large scale, or require very large numbers of parameters, much of my work lies at the intersection of statistics and machine learning.

Keywords (methods)

Bayesian inference, deep learning, density estimation, Gaussian processes, hierarchical models, Monte Carlo

Keywords (applications)

anything with interesting data and decisions, astrophysics, cosmology

photograph of Mark Naylor

Mark Naylor

Research interests

I am interested in how we can use Bayesian methods to generate and test falsifiable forecasts of natural hazards and other geophysical systems. This involves data challenges such as accessing relevant data in times of crises; how to operationalise algorithms which have been developed for retrospective analyses in peace time to be applied during an unfolding crisis; and how to communicate a useful, usable and used message to relevant stakeholders and communities.

Keywords (methods)

Bayesian analysis, point process modelling, INLA, Monte Carlo methods

Keywords (applications)

probabilistic hazard estimation, decision making under uncertainty, hazard mitigation and disaster risk reduction, near-realtime forecasting

photograph of Ken Newman

Ken Newman

Research interests

I am interested in applications to and methodology for ecological and environmental statistics. This includes modelling animal population dynamics, estimation of animal abundance, survival, and fecundity. I'm also interested in sampling methods, including spatial sampling, error correction of land use maps, and approximating complex deterministic models with Gaussian processes.

Keywords (methods)

state-space models, sampling, Bayesian hierarchical models

Keywords (applications)

ecology, fish and wildlife management, environment

photograph of Glenna Nightingale

Glenna Nightingale

Research interests

I am interested in applying statistical principles to real world problems in public health, ecology, and demography. I enjoy data visualization, building R Shiny apps, and teaching statistics through data storyboarding.

Keywords (methods)

Bayesian inference, Natural experiment evaluation, Point process models, Spatial epidemiology, time series models

Keywords (applications)

Behavioural Ecology, biodiversity, demography, public health

photograph of Matthew Nolan

Matthew Nolan

Research interests

Brain circuit mechanisms that implement computations important for cognition. Focus on spatial cognition, high resolution optical imaging of neural circuitry, electrophysiological recording and analysis of single neurons and networks of neurons, modelling and simulation of neural activity.

Keywords (methods)

neural circuit analysis, electrophysiology, neural modelling

Keywords (applications)

neural circuit computation, spatial cognition

photograph of Ugur Ozdemir

Ugur Ozdemir

Research interests

Ugur is a political scientist mainly working on political behavior using survey and social media data. He is using machine learning methods (dimensionality reduction and classification, mainly) in his research. Ugur is interested in methodological challenges and opportunities of new forms of data for social scientific questions.

Keywords (methods)

structural equation modelling, classification, dimensionality reduction, metric learning

Keywords (applications)

political behaviour, survey data, spatial models of elections

photograph of Javier Palarea-Albaladejo

Javier Palarea-Albaladejo

Research interests

Development and application of statistical modelling and data analysis methods in interdisciplinary and collaborative scientific research. Particular interest in compositional data analysis as applied to chemical and behavioural data. Range of experience across animal, environmental and health research, from experimental design to advanced data modelling; externally-funded projects; consulting and training for scientists.

Keywords (methods)

censored and missing data, compositional data analysis, high-throughput data analysis, multivariate data analysis, R programming, statistical modelling

Keywords (applications)

behavioural and time use patterns, bioinformatics, biosciences, chemometrics, vaccine trials

photograph of Yiannis Papastathopoulos

Yiannis Papastathopoulos

Research interests

I am a statistician working at the interface between applied probability and statistical inference. My research focusses on developing statistical models for extreme events through asymptotic characterizations of stochastic processes and random phenomena. This is combined with implementation of Bayesian methods and computationally efficient procedures for statistical inference.

Keywords (methods)

approximate Bayesian inference, spatial statistics and point processes, Markov random fields, Extreme value theory, methods and applications

Keywords (applications)

environment, ocean engineering, pharmaceutical statistics, grid cells, political conflicts

photograph of Richard Parker

Richard Parker

Research interests

I work as a medical statistician in Edinburgh Clinical Trials Unit, and my research interests are primarily in clinical trial design, multiple testing, and studies of agreement (especially Bland-Altman limits of agreement). I apply statistical methods to a varied range of applications in medicine and public health.

Keywords (methods)

agreement method, clinical trial design, end-digit preference, feasibility studies, multiple testing, studies of agreement

Keywords (applications)

biostatistics, clinical trials, medical statistics, medicine, public health

photograph of Lindsay Paterson

Lindsay Paterson

Research interests

My research uses mainly survey data to analyse questions relating to educational expansion. I am particularly interested in the impact of educational institutions on students' learning, and therefore much of my empirical work uses mixed models or more complex forms of multilevel modelling. I am also interested in the impact which education has on other experiences, such as on social mobility or on civic participation.

Keywords (methods)

longitudinal analysis, multilevel modelling, loglinear models, panel data

Keywords (applications)

education, social mobility, civic values

photograph of Daniel Paulin

Daniel Paulin

Research interests

Bayesian computation, high dimensional probability, data assimilation, machine learning and optimization

Keywords (methods)

Markov chain Monte Carlo, Hamiltonian dynamics, concentration inequalities, matrix-free optimization methods

Keywords (applications)

statistical genetics, stochastic volatility models in finance, Weather forecasting

photograph of Jorge Penarrubia

Jorge Penarrubia

Research interests

My research focuses on non-equilibrium gravitational systems, the evolution of gravitating objects embedded in a clumpy environment, and the nature of dark matter. To this aim I use a range of statistical techniques. I am also interested in Bayesian methods of inference and the analysis of large observational data sets.

Keywords (methods)

Bayesian techniques, Monte-Carlo models, multi-dimensional diffusion processes, N-body simulations, stochastic calculus

Keywords (applications)

dark matter halos, galaxies, planetary systems, stellar clusters

photograph of Marcelo Pereyra

Marcelo Pereyra

Research interests

My research is mainly about new mathematical theory, methods and algorithms to solve challenging inverse problems related to mathematical and computational imaging. I am particularly interested in new Bayesian analysis and computation approaches, and in the synergy between modern stochastic simulation, convex optimisation, and machine learning methodology.

Keywords (methods)

Bayesian inference, convex optimisation., high-dimensional statistics, Markov chain Monte Carlo, uncertainty quantification

Keywords (applications)

Computational imaging, remote sensing, medical imaging

photograph of Gareth W. Peters

Gareth W. Peters

Research interests

Statistics and machine learning for risk, insurance and finance, mathematical statistics, computational statistics and stochastic processes.

Keywords (methods)

time series modelling, econometrics, insurance reserve models and capital modelling, spatial temporal modelling and supervised and unsupervised machine learning methodology

Keywords (applications)

risk management, insurance modelling, cyber risk modelling, mortality modelling and demographics, futures modelling, interest rate models, stress testing, portfolio selection and performance assessment, green finance, green bonds, blockchain technology and virtual currencies, regulations research, algorithmic trading and high frequency finance, insurance for extreme weather and climate events

photograph of Chris Ponting

Chris Ponting (CMVM representative)

Research interests

Pinpointing DNA changes that predispose individuals to common genetic disease, and determining how such changes alter gene expression and affect development, cells and organs. We apply and combine methods developed by others to address some of the most important questions in genetics and human disease.

Keywords (methods)

deep learning, Mendelian randomisation, genome-wide association studies

Keywords (applications)

complex disease, cancer, DNA/RNA sequence analysis at all scales from single cells to human subpopulations

photograph of Chris Pooley

Chris Pooley

Research interests

I work on developing fast Bayesian tools and techniques applicable to a broad range of problems. I am developing novel algorithms to help speed up inference and perform fast, reliable model selection. On the applied side I'm looking into incorporating genetics into epidemiology and the analysis of ecological data. I am interested in creating easy to use software to facilitate the widespread use of advanced Bayesian techniques, and I also have a keen interest in deep learning methods, such as convolutional neural networks.

Keywords (methods)

Bayesian inference, capture-recapture, compartmental models, epidemic models, Markov processes, MCMC, mixed models, model selection, stochastic models

Keywords (applications)

ecology, epidemiology, genetics

photograph of Giovanni Rabitti

Giovanni Rabitti

Research interests

My research is focused on methods for the uncertainty quantification and the sensitivity analysis of computer simulation experiments. My interests also include their application to risk analysis models.

Keywords (methods)

Computer experiments, Sensitivity analysis

Keywords (applications)

risk analysis, Actuarial and financial modelling

photograph of Kevin Ralston

Kevin Ralston

Research interests

Survey data to ask and answer questions of sociological/social science interest

Keywords (methods)

Generalized Linear Models, Models for Longitudinal Panel Data, Event History Analysis

Keywords (applications)

Statistics anxiety, young people and work, family and childbearing

photograph of Gail Robertson

Gail Robertson

Research interests

I am an applied statistician at the Statistical Consultancy Unit with a background in quantitative ecology and infectious disease epidemiology. I am involved with the development of data analysis methods in a variety of fields including environmental sciences, forensic statistics, and decision modelling

Keywords (methods)

mixed models, spatial analysis, mapping spatial data, temporal data analysis

Keywords (applications)

epidemiology, ecology, environmental science, forensic statistics, decision modelling

photograph of Gordon Ross

Gordon Ross

Research interests

My research centres around the application of statistical and machine learning methods for predicting time series and point processes. Specific interests include: 1) Detection of abrupt changes/anomalies in time series, 2) Bayesian statistics, particularly nonparametric methods, 3) Risk analysis, including financial markets and earthquake forecasting, 4) Retail analytics, 5) Cyber security.

Keywords (methods)

Bayesian inference, time series, nonparametrics, point process

Keywords (applications)

finance, risk, cybersecurity, earthquakes, seismology

photograph of Xavier Rubio-Campillo

Xavier Rubio-Campillo

Research interests

I'm a computational archaeologist using quantitative methods to explore long-term cultural change. I use a range of statistical tools to tackle current challenges of archaeological research such as the integration of uncertainty within the analysis, the use of multiscalar approaches or spatio-temporal modelling. I have applied these methods to very diverse scenarios including large-scale trade within the Roman Empire, settlement dynamics in Prehistory or battlefield archaeology.

Keywords (methods)

agent-based models, Bayesian inference, Monte-Carlo methods, spatio-temporal analysis, uncertainty quantification

Keywords (applications)

archaeology, long-term adaptation, neolithic dispersion, cultural evolution, history

photograph of Colin Rundel

Colin Rundel

Research interests

My research interests are in Statistical computing, Bayesian methods for spatial statistics, in particular with applications in biology and ecology.

Keywords (methods)

Computing, Gaussian processes, Reproducibility, Spatial Statistics

Keywords (applications)

education, Environmental and ecological modelling

photograph of Sotirios Sabanis

Sotirios Sabanis

Research interests

Analysis of Langevin Monte Carlo sampling algorithms (including the unbiased estimators' version, e.g. SGLD) in a non-Markovian environment, i.e. observation data need not to be Markovian, and their applications in Bayesian learning. Any application of these methodologies to financial data is of keen interest to me.

Keywords (methods)

Bayesian inference, high-dimensional sampling algorithms, mathematical finance, stochastic algorithms/approximations

Keywords (applications)

engineering, portfolio management, risk

photograph of Guido Sanguinetti

Guido Sanguinetti

Research interests

Development of data modelling techniques for biological data, with a particular focus on understanding the dynamics and regulation of gene expression. Stochastic models of chemical reaction networks, reaction-diffusion processes

Keywords (methods)

Bayesian inference, graphical models, chemical master equation

Keywords (applications)

next-generation sequencing, epigenomics, gene expression

photograph of Linus Schumacher

Linus Schumacher

Research interests

I develop mathematical models of cellular processes in biology, especially the collective behaviour of cell populations. I use a variety of tools, including differential equations, agent-based simulations, and stochastic processes, and the main application areas are tissue development, stem cells, and regenerative medicine. As part of these efforts I am interested in data-driven modelling and practical methods for parameter inference and model comparison.

Keywords (methods)

approximate Bayesian inference, agent-based models, hybrid models, master equations, dynamical systems

Keywords (applications)

developmental biology, stem cells, regenerative medicine, cell migration

photograph of Andrew Schurer

Andrew Schurer

Research interests

Understanding the causes of climate change over the last thousand years, in order to improve predictions of the future. My research incorporates running climate models and using the results to study observed climate variability, disentangling the influence of external forcings from internally variability.

Keywords (methods)

Bayesian analysis, Detection and attribution methods, Monte Carlo methods

Keywords (applications)

statistical climatology

photograph of Torben Sell

Torben Sell

Research interests

Statistical methodology, Bayesian statistics, High-dimensional statistics and Statistical learning

Keywords (methods)

Markov chain Monte Carlo, Piecewise deterministic Markov processes, Particle filtering, Function space sampling methods

Keywords (applications)

Stochastic control, Bayesian neural networks

photograph of Sohan Seth

Sohan Seth

Research interests

I run the Data Science Unit (DSU) at the School of Informatics. I am interested in building interpretable models for analysing real-world data with a focus on science, health, people and environment (SHaPE).

Keywords (methods)

Bayesian methods, causal models, computer vision, explainable models, exploratory data analysis, machine learning, model criticism, predictive modelling, time series modelling, unsupervised learning

Keywords (applications)

art, climate, data science, disease characterisation, health informatics, medical diagnosis, patient stratification, remote sensing, risk prediction, social sciences, sustainable development, time-resolved spectroscopy

photograph of Serveh Sharifi Far

Serveh Sharifi Far

Research interests

Identifiability problem in statistical models and parameter redundancy, estimating population size in hard-to-reach populations, Statistics and Mathematics pedagogy.

Keywords (methods)

structural equation modelling, parameter redundancy, multiple systems estimation

Keywords (applications)

education, social sciences

photograph of Valeria Skafida

Valeria Skafida (CAHSS representative)

Research interests

Exploring and understanding changes in children's health and wellbeing using longitudinal data with a focus on social inequalities and with reference to relevant social policy. Areas of specific interest include infant diet and child nutrition; domestic violence and children's wellbeing.

Keywords (methods)

survival analysis, longitudinal analysis, fixed and random effects models, GLM, dimension reduction (principal component analysis)

Keywords (applications)

public health, epidemiology, social policy, sociology

photograph of Athina Spiliopoulou

Athina Spiliopoulou

Research interests

I am a health data scientist with background in machine learning. I study how genetic factors and biomarkers (e.g. lipids measured in blood) contribute to disease. A key component of my work is the analysis of large-scale, high-dimensional datasets, that link genetic and biomarker measurements with electronic health records.

Keywords (methods)

Bayesian inference, feature engineering, Mendelian randomisation (causal inference using genetic markers), mixtures of experts, shared latent spaces, shrinkage models, variable selection

Keywords (applications)

autoimmune diseases, biomarker discovery and validation, disease endotyping, disease risk stratification, genetic epidemiology, precision medicine

photograph of David Sterratt

David Sterratt

Research interests

To understand aspects of neurobiological systems I use deterministic and stochastic simulations, data analysis and parameter inference methods. I develop methods and software in R and python to support: (a) multilevel simulation of electrical and biochemical activity in neurons; and (b) transformation, visualisation, and analysis of images of retinae. I am active in data science and neuroscience education.

Keywords (methods)

Bayesian inference, data science, dynamical systems, stochastic models, teaching

Keywords (applications)

education, neuroscience, retinal imaging

photograph of Amos Storkey

Amos Storkey

Research interests

I work in the School of Informatics. Although by no means exhaustive, my research covers the following areas: machine learning, deep learning and neural networks, transfer Learning, Bayesian methods, machine learning markets. And my research has applications in medical imaging, health, finance and music.

Keywords (methods)

Bayesian inference, Bayesian networks, game theory, meta-learning, neural networks, stochastic differential systems, time-series

Keywords (applications)

brain imaging, CT, diffusion tensor imaging, epidemiology, financial models, fMRI, MRI, music generation, retinal imaging

photograph of George Streftaris

George Streftaris

Research interests

Bayesian statistical modelling and inference, focussed on the development and application of novel methodology used in stochastic processes for partially observed populations. I am particularly interested in inter-disciplinary work involving statistical methodology applied in infectious epidemiology, actuarial mathematics and life sciences.

Keywords (methods)

Bayesian inference, epidemic models, Markov chain Monte Carlo, model assessment, predictive modelling

Keywords (applications)

infectious epidemiology, actuarial mathematics, morbidity, life sciences

photograph of Lukasz Szpruch

Lukasz Szpruch

Research interests

I have a broad research interest in probability theory, mean-field models, stochastic control, statistics and quantitative finance. I am also researching on deep neural networks and reinforcement learning. I have worked with the financial services industry on topics such as model calibration or risk computation.

Keywords (methods)

machine learning, Markov Chain Monte Carlo methods, mean field models

Keywords (applications)

cyber security, energy markets, financial services, insurance

photograph of Simon Taylor

Simon Taylor

Research interests

My main area of research is currently around changepoint detection for periodic data, which has applications in public health, finance and the environment. Other interests include the analysis of eye-tracking data, computational intensive methods and Bayesian model selection.

Keywords (methods)

Bayesian inference, Computational intensive methods, Sequential Monte Carlo, Changepoint detection, Time series analysis

Keywords (applications)

biostatistics, geoscience, health

photograph of Albert Tenesa

Albert Tenesa

Research interests

The overarching aim of my lab is to understand what drives phenotypic variation in populations. To achieve this aim, we develop computational and statistical tools to analyse genetic and genomic data at scale. We also combine lifestyle and environmental risk factors with genetic risk factors for the prediction of disease risk or complex traits. @GroupTenesa

Keywords (methods)

machine learning, mixed linear models, generalized mixed models, high performance computing, regularization methods

Keywords (applications)

genomics, quantitative genetics, genetic epidemiology, animal breeding, population genetics, statistical genetics

photograph of Dusan Uhrin

Dusan Uhrin

Research interests

Development and application of Nuclear Magnetic Resonance (NMR) spectroscopy to the structure elucidation and reaction kinetics of molecules in solution. NMR and mass spectrometry (MS) analysis of complex mixtures such as beverages, biological extracts or environmental matrices. Use of multivariate analysis for the interpretation of NMR and MS spectra of complex mixtures.

Keywords (methods)

mass spectrometry, methodology of NMR spectroscopy, multivariate analysis

Keywords (applications)

complex carbohydrates, complex mixtures analysis, dissolved organic matter, metabolites, molecular structure, reaction monitoring, Scotch Whisky

photograph of Catalina Vallejos

Catalina Vallejos

Research interests

Development, implementation and application of novel Bayesian methodology that is motivated by complex biomedical problems and data types (e.g. those generated by cutting-edge genomic technologies). Development of statistical models that are robust to measurement noise and other sources of unwanted heterogeneity. Implementation of open-source analysis tools

Keywords (methods)

Bayesian inference, hierarchical models, survival analysis, competing risks

Keywords (applications)

biomedicine, single cell transcriptomics, electronic health records

photograph of Tod Van Gunten

Tod Van Gunten

Research interests

Economic and political sociology, social network analysis, computational social science

Keywords (methods)

Dimension reduction, networks, multilevel modeling

Keywords (applications)

political networks, economic networks, formal analysis of belief systems

photograph of Antonio Vergari

Antonio Vergari

Research interests

Efficient and reliable machine learning in the wild, tractable probabilistic modeling, combining learning and complex reasoning

Keywords (methods)

tractable inference, probabilistic graphical models, robust estimation

Keywords (applications)

data understanding, model selection, constraint learning and reasoning, biomed AI

photograph of Gil Viry

Gil Viry

Research interests

Gil is a Lecturer in Sociology, co-leads the Social Network Analysis in Scotland (SNAS) Research Group and is a member of Edinburgh Q-Step team for promoting statistical methods within the social sciences. His substantive research focuses on spatial mobility, social networks, family and personal life. He has a keen interest in studying the spatiality of social networks and how physical distance and mobility behaviours relate to individuals' social and professional integration over the life course.

Keywords (methods)

social network analysis, social survey methods, sequence analysis

Keywords (applications)

spatial mobility, space, family, life course

photograph of Sara Wade

Sara Wade

Research interests

My research interests span Bayesian inference and machine learning. I am particularly interested in nonparametric methods for regression, clustering, and dimension reduction, and computational algorithms for full posterior inference, as well as fast algorithms for approximate inference, based on variational methods and maximum aposteriori estimates.

Keywords (methods)

Bayesian inference, machine learning, clustering, nonparametrics, density regression, Dirichlet process, Gaussian process

Keywords (applications)

medicine, imaging data, time-to-event data, Alzheimer's disease

photograph of Christopher Weir

Christopher Weir

Research interests

I am one of the leads of the statistics team in Edinburgh Clinical Trials Unit, with a particular interest in early phase clinical trial design and health services research / implementation studies. I collaborate with clinical investigators to develop and deliver clinical trials and health research studies, alongside development of novel clinical trials methodology. Specific research interests have included practical implementation of Bayesian adaptive designs in clinical trials and development of statistical methodology to enable the evaluation of potential surrogate outcome measures for clinical ...

Keywords (methods)

Bayesian adaptive clinical trial design, cluster-randomised and stepped-wedge trials, surrogate outcome evaluation

Keywords (applications)

clinical trials, medicine

photograph of Chris Williams

Chris Williams

Research interests

I am interested in a wide range of theoretical and practical issues in machine learning, statistical pattern recognition, probabilistic graphical models and computer vision. This includes theoretical foundations, the development of new models and algorithms, and applications.

Keywords (methods)

Gaussian processes, time series understanding, condition monitoring, scene understanding, messy data, unsupervised learning

Keywords (applications)

image and scene analysis, intensive care unit monitoring, data science for science

photograph of Amy Wilson

Amy Wilson

Research interests

I am an applied statistician interested in statistical modelling for applications in energy and the law. I have worked on a variety of applied problems. These have included assessing the risk of long-term electricity outages, quantifying uncertainty in energy policy models, evaluating the strength of forensic evidence (for e.g. drug traces on banknotes) and quantifying uncertainty in the outcome of civil legal cases.

Keywords (methods)

Bayesian inference, statistics of extremes, uncertainty quantification, time series modelling, statistical inference

Keywords (applications)

engineering, energy policy, forensic science, law, electricity shortfalls

photograph of Andrea Wilson

Andrea Wilson

Research interests

Development of statistical and mathematical models and computational tools to determine how the genetics of individuals and diverse non-genetic factors together influence the dynamics of infectious diseases and their impact on the health and performance of individuals and of entire livestock populations.

Keywords (methods)

Bayesian inference, hierarchical mixed models, statistical genetics, random regression models, ordinary differential equations, stochastic modelling of non-linear dynamic processes, social network analysis

Keywords (applications)

infection dynamics, disease transmission, quantitative genetics applied to livestock populations, individual and group aggression, resilience

photograph of Maria Wolters

Maria Wolters

Research interests

In my work, I leverage data and technology to support people with chronic conditions in living rich and meaningful lives. I have expertise in evaluating services, apps, and products, and I am particularly interested in helping people take control of and make sense of their own data.

Keywords (methods)

missing data, human-data interaction, mixed qualitative and quantitative methods, questionnaire design and analysis, evaluation

Keywords (applications)

medicine, design, sports, wellbeing

photograph of Simon Wood

Simon Wood

Research interests

Computational and applied statistics. Especially stable, efficient and scalable computation methods for regression models based on smooth functions of predictors (GAMs, quantile additive models, location scale models, functional data models and the like). Penalized, empirical Bayes and full Bayes methods. Multidimensional smoothing methods. Applications in energy demand forecasting, fisheries assessment, epidemiology and ecology. Author of mgcv package in R.

Keywords (methods)

fast stable computation, MGCV, penalized regression, Smoothing

Keywords (applications)

energy, environment

photograph of Bruce Worton

Bruce Worton

Research interests

My interests are primarily in research involving statistical modelling of challenging problems and related statistical inference techniques. I have undertaken collaborative research with Scotland's Rural College (SRUC) and the James Hutton Institute, Dundee, concerned with modelling the movements of larvae using flexible diffusion processes to characterize the behaviour of larvae when exposed to attractant/repellent compounds

Keywords (methods)

statistical modelling, applied statistics, bootstrap computation, likelihood inference, efficient Bayesian inference

Keywords (applications)

biology, ecology, entomology, geosciences

photograph of Nick Wray

Nick Wray

Research interests

Industrial applications (locomotive performance monitoring), streamflow, land use and climate change

Keywords (methods)

Dickey-Fuller etc, Mann-Kendall, Pettitt, time series, wavelets

Keywords (applications)

Time series Analysis, Trend and Discontinuity Analysis