STAT40380 Bayesian Analysis

Academic Year 2019/2020

Bayesian Analysis models all unknown quantities in a coherent probabilistic framework. Full probability distributions for model parameters conditional on observed data are derived. This module explores how this can be done, both algebraically and computationally. Understanding the Bayesian approach to inference is central and manipulation of conditional distributions is key. The free software package JAGS will be used to perform analysis on a range of statistical models, from simple to complex hierarchical models. Topics covered include: conditional probability, Bayes' Theorem, prior distributions, conjugacy, the likelihood principle, multi-parameter problems, Bayesian hypothesis testing and model checking, methods for finding the posterior mode, Markov Chain Monte Carlo, advanced Bayesian modelling. Illustrative examples from the scientific literature will be used.

Show/hide contentOpenClose All

Curricular information is subject to change

Learning Outcomes:

By the end of the course the students should be able to propose and fit a fully Bayesian statistical model to a wide variety of data sets. They should be able to check the model and give a critique of the Bayesian process as opposed to its Frequentist counterpart.

Student Effort Hours: 
Student Effort Type Hours
Tutorial

5
Computer Aided Lab

5
Specified Learning Activities

38
Autonomous Student Learning

42
Online Learning

24
Total

114
Approaches to Teaching and Learning:
Video lectures twice per week via the VLE. Physical tutorials in smaller groups to work on practical problems face-to-face, and computer labs with demonstrator to work through coding examples. Solutions to tutorial and lab sheets posted after each class as well as worked solutions to assignments. 
Requirements, Exclusions and Recommendations
Learning Requirements:

You should have completed a basic course in statistics including probability, inference, hypothesis testing, estimation and regression.


Module Requisites and Incompatibles
Incompatibles:
STAT40390 - Bayesian Analysis


 
Assessment Strategy  
Description Timing Open Book Exam Component Scale Must Pass Component % of Final Grade
Examination: End of semester examination 1 hour End of Trimester Exam No Graded No

50
Continuous Assessment: Assignments will be a mix of theory and computer based problem sheets. Two minor assignments worth 5% each and two major worth 20% each. Throughout the Trimester n/a Graded No

50

Carry forward of passed components
No
 
Resit In Terminal Exam
Autumn Yes - 2 Hour
Please see Student Jargon Buster for more information about remediation types and timing. 
Feedback Strategy/Strategies

• Feedback individually to students, post-assessment
• Group/class feedback, post-assessment

How will my Feedback be Delivered?

Not yet recorded.

Bayesian Statistics: An Introduction by Lee
Bayesian Data Analysis by Gelman, Carlin, Stern, and Rubin.
Name Role
Dr Riccardo Rastelli Lecturer / Co-Lecturer
John O'Sullivan Tutor