International Workshop on Bayesian Data Analysis and Software Application

Rationale
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.

Workshop Objectives
The workshop aims to:

  •  Introduce the participants smoothly into Bayesian statistical methods, from basic concepts to hierarchical models, model building and model evaluation.
  •  Introduce students to applying Bayesian approach to modelling locally relevant public health challenges.

Background to the course
The course will provide participants with numerous bio-statistical examples (e.g. survival analysis, meta-analyses, longitudinal studies including growth curve modelling, spatial modelling, etc.), which will be used to illustrate the theoretical Bayesian concepts.
The course is scheduled for classroom teaching and computer exercises, and uses the software packages WinBUGS, OpenBUGS, JAGS and also their interfaces with R such as R2WinBUGS, R2OpenBUGS and rjags.
The course is based on the Wiley book written by Prof Emmanuel Lesaffre and Lawson (2012), entitled Bayesian Biostatistics.
Participants are expected to attend all sessions. This workshop will be offered in English.
Eligibility
The course assumes a good knowledge of regression techniques (linear, logistic, etc.) and some knowledge on models for correlated data. Some programming skills are essential, and experience with R is desirable. Participants are advised to familiarize themselves in advance with the basics of R and RStudio.

Main Instructor
Prof. Emmanuel Lesaffre

COURSE CONTENT

 

Day 1

MORNING

Session 1: Classical frequentist statistics and the likelihood function
Session 2: Bayes theorem, posterior distribution and sampling from posterior distribution

AFTERNOON

Session 3: Posterior summary measures, posterior predictive distribution and Bayesian hypothesis testing
Practical session: Use of R to compute posterior distribution and posterior summary measures

 

            Day 2

MORNING

Session 4: More than one parameter - Bayesian linear regression
Session 5 (part I): Introduction to MCMC techniques

AFTERNOON

Session 5 (part II): Introduction to MCMC techniques
Practical session: Introduction to Win/OpenBUGS

 

            Day 3

MORNING

Session 6: Choosing the prior distribution
Session 7: Bayesian hierarchical models

AFTERNOON

Session 8: Model selection and model checking
Practical session: Introduction to R2Win/R2OpenBUGS

 

            Day 4

MORNING

Session 9: Meta analyses + practical exercises

AFTERNOON

Session 10: Survival analysis + practical exercises

 

            Day 5

MORNING

Session 11: Spatial models + practical exercises

AFTERNOON

Session 12: Longitudinal analysis + practical exercises

 

Note: Deadline for application of the Bayesian Workshop has been extended to 31st August 2018

For detailed course contents please click here.

International Workshop on Bayesian Data Analysis and Software Application.

1st - 15th October, 2018