Bayesian Analysis of Small Area Health Data using WinBUGS/GeoBUGS Workshop

Start date: 
Wednesday, November 7, 2012 - 20:00 to Friday, November 9, 2012 - 23:45
Event Speaker: 
Professor Nicky Best and Dr Marta Blangiardo
Location: 
Imperial College, London
United Kingdom
Use of Bayesian methods is becoming increasingly widespread in spatial epidemiology for analysing data with complex structure – for example, hierarchical data, spatial and temporal dependence, measurement error and so on. However, very few data analysts have any formal training in Bayesian methods. This three-day course aims to introduce quantitative researchers to the basic principles of Bayesian inference and simulation-based methods for estimating Bayesian models, and to then focus specifically on Bayesian methods for disease mapping and modelling of small area health data, and how to implement these methods using the WinBUGS/GeoBUGS software. There is a large practical component to this course with time for hands-on data analysis using a variety of examples in spatial epidemiology. 
 

Course content: 

  • Basic principles of Bayesian inference 
  • Introduction to Monte Carlo simulation methods and MCMC
  • Introduction to spatial data analysis (area and point data, spatial dependence, tests for overdispersion and autocorrelation)  
  • Introduction to Bayesian hierarchical models for count and continuous data
  • Bayesian spatial models for small area health data  
  • Ranking, classification and detection of unusual areas
  • Ecological and hierarchical related regression models
  • Bayesian space-time models for small area health data
  • Hands-on experience of using WinBUGS/GeoBUGS software software to fit Bayesian models for small area health data

 

Target audience: 

Statisticians, data analysts and quantitative health researchers who are interested in learning about Bayesian methods for modelling small area health data, and how to implement these methods using the WinBUGS/GeoBUGS. No previous experience of Bayesian methods or WinBUGS is necessary. However, attendees should have strong quantitative data analysis skills, including a good working knowledge of the practice and principles of linear and generalized linear models, and be familiar with standard statistical terminology and common probability distributions (normal, binomial, Poisson). 

 

Course tutors:

This three day specialist workshop will be led by researchers from the MRC-HPA Centre for Environment and Health:

 

Professor Nicky Best

Nicky Best is Professor of Statistics and Epidemiology at Imperial College London. Her research involves development and application of Bayesian statistical methods in health and social science with particular interest in spatial epidemiology and the development and application of Bayesian hierarchical models to model small area health data.

 

Dr Marta Blangiardo

Marta Blangiardo is a lecturer in Biostatistics at Imperial College London. Her key research areas are: Bayesian modelling of exposure to air pollutants, integration of air pollution concentration from different sources; Bayesian hierarchical models for combining individual and ecological data in epidemiological studies; Bayesian measurement error models.

 

Laptops and Software download:

Participants are asked to bring their own laptops for the practical classes if possible. We ask that you work in pairs during the practical classes, so if you do not have a laptop we can always pair you up with someone that does.

If you intend to bring your own laptop, it would save time if you could download and install the software beforehand.

 

You will need to install WinBUGS version 1.4.3, which can be downloaded from the WinBUGS website:

http://www.mrc-bsu.cam.ac.uk/bugs/winbugs/contents.shtml

 

This site also contains instructions for installation. You will need to download and install the main WinBUGS 1.4 package, and then the patch to upgrade this to 1.4.3 and the key file to allow unrestricted use.  

 

If you are coming from outside the UK, please remember to bring an adaptor plug so that you can plug your laptop into UK power sockets!

 

Early registration is recommended as places for the workshop are limited to between 20 - 25 participants and are allocated on a first come first served basis. 

 

The registration fee is £360. 

 

Book your place on the Bayesian Analysis of Small Area Health Data Using WinBUGS/GEOBUGS workshop online.