Applied Bayesian Modelling - MA538

Location Term Level Credits (ECTS) Current Convenor 2019-20
Canterbury Autumn
View Timetable
6 15 (7.5) DR F Leisen

Pre-requisites

Prerequisite module: MAST5007 Mathematical Statistics

Restrictions

None

2019-20

Overview

The origins of Bayesian inference lie in Bayes' Theorem for density functions; the likelihood function and the prior distribution combine to provide a posterior distribution which reflects beliefs about an unknown parameter based on the data and prior beliefs. Statistical inference is determined solely by the posterior distribution. So, for example, an estimate of the parameter could be the mean value of the posterior distribution. This module will provide a full description of Bayesian analysis and cover popular models, such as the normal distribution. Initially, the flavour will be one of describing the Bayesian counterparts to well known classical procedures such as hypothesis testing and confidence intervals. Outline Syllabus includes: Bayes Theorem for density functions; Exchangeability; Choice of priors; Conjugate models; Predictive distribution; Bayes estimates; Sampling density functions; Gibbs samplers; OpenBUGS; Bayesian hierarchical models; Applications of hierarchical models; Bayesian model choice.

Details

This module appears in:


Contact hours

36 hours

Method of assessment

80% Examination, 20% Coursework

Indicative reading

A. Gelman, J.B. Carlin, H.S. Stern, D.B. Dunson, A. Vehtari and D.B. Rubin (2014). Bayesian Data Analysis. 3rd Edition, Chapman & Hall/CRC Texts in Statistical Science
D. Gamerman and H.F. Lopes (2006). Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference. 2nd Edition, Taylor and Francis.

See the library reading list for this module (Canterbury)

Learning outcomes

The intended subject specific learning outcomes. On successfully completing the module students will be able to:

1 derive posterior distributions when analytically tractable;
2 derive posterior summaries, such as estimates, from the posterior distribution, including the predictive distribution;
3 construct Bayesian hierarchical models and implement them in a suitable software package;
4 critically evaluate software output using convergence diagnostics;
5 interpret and report the output for inferential purposes.

The intended generic learning outcomes. On successfully completing the module students will be able to:

1 use a logical mathematical approach to solve problems;
2 work with relatively little guidance;
3 solve problems and communicate in writing more effectively.

University of Kent makes every effort to ensure that module information is accurate for the relevant academic session and to provide educational services as described. However, courses, services and other matters may be subject to change. Please read our full disclaimer.