Applied Bayesian Methods

Course code STAT3120Units 10Level 3000Faculty of Science and Information TechnologySchool of Mathematical and Physical Sciences

The course introduces students to Bayesian thinking and methods from an applied point of view; covering the use of prior information, Bayes' rule and inference in standard situations such as proportions, means and relationships between variables. An applied view on Markov chain Monte Carlo methods will also be given. These methods are becoming popular among applied statisticians and analysts from disciplines such as, Economics, Quantitative finance, Health, Environmental science, Engineering and other applied areas, especially because prior information can be incorporated directly into analyses in a sensible way.

This course is open to students in the BMath program (including double degree programs) or to students in other programs who have received explicit permission from the Head of Discipline of Statistics.

This course is shared by the Universities of Newcastle, Western Sydney and Wollongong as part of the Applied Statistics Education and Research Collaboration (ASEARC). In some years, the course will be beamed live from one of the other institutional partners using the Access Grid Room rather than being taught face-to-face at Newcastle.

Available in 2015

Callaghan CampusSemester 2
Previously offered in 2014
ObjectivesOn successful completion of this course, students will be able to:

1. understand Bayesian thinking;
2. use prior information and Bayes' rule in probability and statistical inference problems;
3. apply Bayesian inference methods to common parameters (binomial, Normal) and to relationships between variables; and
4. compare these with frequentist methods.
ContentIntroduction to Bayesian thinking

The use of prior information

Bayesian estimation of:
- the binomial parameter
- the Normal mean and variance
- the poisson parameter

Empirical Bayes estimation

Bayesian estimation in:
- analysis of variance
- regression

Markov chain Monte Carlo methods
TransitionNot applicable.
Industrial Experience0
Assumed KnowledgeSTAT2010
Modes of DeliveryExternal Mode
Internal Mode
Teaching MethodsLecture
Computer Lab
Assessment Items
Essays / Written Assignments
Examination: Formal
Contact HoursLecture: for 2 hour(s) per Week for Full Term
Computer Lab: for 2 hour(s) per Week for Full Term
Compulsory Components
Requisite by EnrolmentThis course is open to students in the BMath program (including double degree programs) or to students in other programs who have received explicit permission from the Head of Discipline of Statistics.
Timetables2015 Course Timetables for STAT3120