Bayesian Analysis Made Easy
Bayesian methods have become popular as a useful tool for data analysis and decision making. Modern software has made this possible and the methods are now applied in a wide range of scientific application areas from medicine to ecology.
This course is aimed at those who are new to Bayesian statistics and want to develop an understanding and application of the methods. Emphasis will be on practical data analysis and interpretation. Only essential theory will be outlined.
The OpenBUGS package, a widely used specialist software for Bayesian analysis, will be used in the presentations as well as other software as appropriate.
The course will include an introduction to and practical exercises in OpenBUGS. The R software will also be used as a graphical tool. In addition to OpenBUGS, during the hands-on computer sessions participants will be able to choose from the statistics packages SAS and Stata.
Who Should Attend?
Scientists and technologists who want an introduction to Bayesian methods for data analysis. No prior knowledge of Bayesian statistics is required. A working knowledge of linear and generalised linear models and statistical distributions is required.
No previous experience of using OpenBUGS or R is required.
How You Will Benefit
By the end of the course, you will have a firm understanding of Bayesian methods and their flexibility. You will also have acquired a working knowledge of specialised software for Bayesian data analysis and will be able to fit and interpret linear and generalised linear models.
What Do We Cover?
- The Bayesian approach to statistics; prior and posterior distributions
- MCMC methods and diagnostics
- Using the OpenBUGS software interactively, using scripts
- Fitting linear and generalised linear models, output and interpretation, model selection
- Analysis of designed studies
- Questions that classic statistics find difficult or cannot answer.
This course has practical exercises written for: OpenBUGS, R, SAS, Stata
The practical exercises use mostly OpenBUGS, supplemented by R where required. SAS or Stata may also be used instead of R and participants will have the opportunity to use SAS or Stata for fitting Bayesian models.