This module will introduce Bayesian analysis with emphasis on data modelling and computational methods. After an overview of foundational concepts in probability theory, students will be introduced to the basic concepts in Bayesian analysis including prior specification, posterior inference prediction and model selection. Monte Carlo methods will be used to approximate quantities of interest. All the important concepts and methods will be explained via examples using advanced statistical software (e.g., R).
Probability theory
Review of Bayes’ theorem and basic probability theory. Monte Carlo methods. Likelihood principle. Subjective probability.
Probabilistic modelling
Conjugate models (e.g. Beta-Binomial, Gamma-Poisson). Prior specification. Posterior inference and prediction. Bayes factor. Linear regression.
Simulation methods
Markov chain Monte Carlo algorithms (e.g. Metropolis-Hastings, Gibbs sampler).
Software packages
Data analysis using advanced statistical packages (e.g. R, RStan).
Lectures supported by data analysis sessions and the use of statistical software packages.
Module Content & Assessment | |
---|---|
Assessment Breakdown | % |
Other Assessment(s) | 100 |