Programme
Section outline
-
The course will run every day from 9h (9h45 on Monday) to 12h30 and then from 13h30 to 17h, and will include two coffee breaks per day.
Monday 5 September:
9:45 - 10:30 Introduction to Bayesian thinking
- frequentist vs bayesian statistics
- conditional probabilities
- Bayes's theorem
10:30 - 11:00 coffee break 11:00 - 12:30 Exercises 12:30 - 13:30 lunch break 13:30 - 15:15 Single-paramater models
- Prior, likelihood and posterior
- some useful distributions
- Bayesian inference
- examples of single parameter models
15:15 - 15:45 coffee break 15:45 - 17:00 Exercises Tuesday 6 September:
9:00 - 10:30 Multiparameter models
- Multiparameter model
- prior/likelihood/posterior
- inference
- Bayes' factor
- model checking
10:30 - 11:00 coffee break 11:00 - 12:30 Exercises 12:30 - 13:30 lunch break 13:30 - 15:15 Hierarchical models
- Hierarchical structure
- exchangeability
- inference
- example and applications
15:15 - 15:45 coffee break 15:45 - 17:00 Exercises Wednesday 7 September:
9:00 - 10:30 Bayesian computations
- Normal approximation
- Monte Carlo methods
- inversion sampling
- rejection sampling
- importance sampling
10:30 - 11:00 coffee break 11:00 - 12:30 Exercises 12:30 - 13:30 lunch break 13:30 - 15:15 Bayesian computations (cont.)
- Markov chains
- introduction to MCMC
15:15 - 15:45 coffee break 15:45 - 17:00 Exercises Thursday 8 September:
9:00 - 10:30 MCMC
- Gibbs sampling
- Metropolis sampling
- Metropolis-Hastings sampling
- burn-in period
- thinning
10:30 - 11:00 coffee break 11:00 - 12:30 MCMC in practice
- convergence
- methods of diagnostic
12:30 - 13:30 lunch break 13:30 - 14:15 Rstan 14:15 - 15:15 Exercises and practical use of Rstan 15:15 - 15:45 coffee break 15:45 - 17:00 Exercises and practical use of Rstan (cont.)