This course is designed to teach students how to use the WinBugs package to carry out probablistic and statistical analyses. It will be based on lectures and practical computer based sessions, giving the students hands-on experience of WinBugs. The students will learn how to write probabilistic models in WinBugs and carry out the process of fitting them to data, as well as how to avoid some of the traps and pitfalls that commonly occur in MCMC model fitting. The course will be aimed at higher level undergraduate and postgraduate students in mathematics and statistics. Some basic knowledge of Bayesian inference will be an advantage, although it is not a prerequisite. The course will be taught by Andrew Thomas (who wrote the WinBugs package), and Bob O'Hara (who's sole contribution has been to complain about the gibberish in the trap messages).
As the number of places on the course is limited, students should pre-register by emailing Bob O'Hara
Dates: Wednesday 23rd -Friday 25th April and Monday 28th - Tuesday 29th April
Times: 10am - 4pm
Place: Mathematics computer room
Teachers: Andrew Thomas and Bob O'Hara, Rolf Nevanlinna Institute
Credits: Students will gain one credit for attending the course
Excelling even myself, here are a couple of .pdfs for the lectures. Of course, the lectures will probably change by the time I give them:
Introduction to Bayesian Analysis
Warning: This will be an immensely boring course, of interest only for lunatics and those believing themselves to be incurable insomniacs.
Here is a provisional outline of the course, with a cat:
And finally, a link to the WinBugs website: go here.