EPM5013 - Bayesian statistical methods - 2018

6 points, SCA Band 2, 0.125 EFTSL

Postgraduate - Unit

Refer to the specific census and withdrawal dates for the semester(s) in which this unit is offered.

Faculty

Medicine, Nursing and Health Sciences

Organisational Unit

Department of Epidemiology and Preventive Medicine

Chief examiner(s)

Professor Andrew Forbes

Coordinator(s)

Associate Professor Lyle Gurrin

Unit guides

Offered

Alfred Hospital

Clayton

  • Second semester 2018 (Online)

Prerequisites

EPM5002, EPM5003, EPM5004, EPM5009, EPM5014, MPH5040.

Co-requisites

Must be enrolled in course code : 3420, 3421, 3422, M6025.

Prohibitions

This unit is only available to students enrolled in the Graduate Certificate, Graduate Diploma or Masters of Biostatistics.

Synopsis

This unit provides a thorough introduction to the concepts and methods of modern Bayesian statistical methods with particular emphasis on practical applications in biostatistics. Comparison of Bayesian concepts involving prior distributions with classical approaches to statistical analysis, particularly likelihood based methods. Applications to fitting hierarchical models to complex data structures via simulation from posterior distributions using Markov chain Monte Carlo techniques (MCMC) with the WinBUGS software package.

Outcomes

Upon successful completion of this unit, students should be able to:

  1. Explain the logic of Bayesian statistical inference i.e. the use of full probability models to quantify uncertainty in statistical conclusions.
  2. Develop and analytically describe simple one-parameter models with conjugate prior distributions and standard models containing two or more parameters including specifics for the normal location-scale model.
  3. Appreciate the role prior distributions and have a thorough understanding of the connection between Bayesian methods and standard 'classical' approaches to statistics, especially those based on likelihood methods.
  4. Recognise situations where a complex biostatistical data structure can be expressed as a Bayesian hierarchical model, and specify the technical details of such a model.
  5. Explain and use the most common computational techniques for use in Bayesian analysis, especially the use of simulation from posterior distributions based on Markov Chain Monte Carlo (MCMC) methods, with emphasis on the practical implementation of such techniques in the WinBUGS package.
  6. Perform practical Bayesian analysis relating to health research problems, and effectively communicate the results.

Assessment

  • Written assignments (80%)
  • Practical exercises (20%)

This unit applies to the following area(s) of study

Additional information on this unit is available from the faculty at: