module 16: bayesian data analysis

module 16: bayesian data analysis

  • module 16: bayesian data analysis
    16/09/2019 - 17/09/2019
    9:30 am - 5:00 pm
    Mon 16/09: 9.30 am - 5.00 pm | S9 room 3.1
    Tue 17/09: 9.30 am - 5.00 pm | S9 room 3.1

Course details

week 2


In the last two decades, the Bayesian approach has become increasingly popular in virtually all application areas. The approach is especially known for its capability to tackle complex statistical modeling tasks. The aim of this course is to introduce the participants smoothly into Bayesian statistical methods, from basic concepts to hierarchical models, model building and model testing. Especially a variety of examples (but mainly in the medical area) will illustrate the theoretical concepts. The course is scheduled into classroom teaching and computer exercises, and uses the software packages WinBUGS and OpenBUGS but also their interfaces with R making use of R2WinBUGS and R2OpenBUGS. The course is partly based on the Wiley book of Lesaffre and Lawson, published in 2012 and entitled: Bayesian Biostatistics.
Learning outcomes

  • Understanding the basic concepts and tools of the Bayesian approach
  • Working with Win/OpenBUGS and R2Win/OpenBUGS for performing Bayesian analyses

Course structure

Day 1, morning

  • A quick reflection on the frequentist approach
  • Bayes theorem and the posterior distribution
  • Posterior summary measures: mean, median, SD, credible interval
  • The posterior predictive distribution
  • Independent sampling of the posterior
Day 1, afternoon
  • Multivariate posterior distributions
  • Independent sampling of the multivariate posterior MCMC sampling: Gibbs Sampling
  • Exercises on using WinBUGS and OpenBUGS
Day 2, morning
  • Choosing the prior distribution
  • Checking and accelerating the convergence of the MCMC procedure
  • Selecting and checking the Bayesian model
  • Exercises on using Win/OpenBUGS and R2Win/OpenBUGS
Day 2, afternoon
A selection of examples from:
  • Application topic 1: Analysis of clustered data
  • Application topic 2: Clinical trials
  • Application topic 3: Survival analysis


Statisticians with a good statistical background, who have programming skills preferably in R.


Emmanuel Lesaffre is Professor of Biostatistics at the KU Leuven and U Hasselt in Belgium. He also holds a honorary professorship at the university of Erasmus, Rotterdam, the Netherlands. He studied mathematics at the University of Antwerp, and received his Doctorate of Science at KU Leuven in 1986. The statistical research of Dr. Lesaffre deals with hierarchical and clustered data with a focus on longitudinal studies, interval censoring, missing data problems, Bayesian methods and in general statistical methods in clinical trials. He has worked in a great variety of medical applications, with focus on research in oral health, cardiology and nursing studies. He is the (co)-author of more than 400 peer reviewed papers in major statistical and clinical journals, and is the (co)-author/editor of nine books.
Dr. Lesaffre is the founding chair of the Statistical Modeling Society, and is a past-president of the International Society of Clinical Biostatistics (2006-2008). In addition he started up a bi-annual international conference on statistical methods in oral health. He is one of the three founding editors of Statistical Modeling and has been Associate Editor of Biometrics, and is currently Associate Editor of Biostatistics. He was elected Fellow of the American Statistical Association, ISI and is an honorary member of the Society of Clinical Biostatistics. Dr. Lesaffre started up the Biostatistical Centre, a predecessor of the Leuven Institute for Biostatistics and statistical Bioinformatics. He was also chair of the department of Biostatistics at Erasmus MC from 2007 to 2014.


Venue Phone: +32 9 264 48 84

Venue Website:

Krijgslaan 281, Gent, 9000, Belgium