Bayesian Methods for Medical Statistics

Description

The course introduces students to the basic principles of Bayesian statistics that have applications in the field of medicine.

Availability

Distance Education - Callaghan

  • Semester 1 - 2017

Learning Outcomes

On successful completion of the course students will be able to:

1. Understand Bayes' rule, prior distributions and their applications to medical statistics;

2. Be able to apply Bayesian methods to medical data using SAS and WINBUGS software;

3. Interpret credible intervals and probabilities from a Bayesian standpoint

4. Impute data using MCMC Bayesian methodology

Content

Topics covered in this course will include the principles of prior and posterior distributions, Bayes' rule for statistical inference, conjugate and non-conjugate priors, the Gibbs sampler, Wishart and inverse Wishart distributions, the Metropolis and Metropolis-Hastings algorithms, Jeffries invariant prior, hierarchical linear models from a Bayesian perspective, the concept of  credible intervals, exchangeable prior models for robust inference, Bayesian mixture models and Markov Chain Monte Carlo (MCMC) models.

 

 

Requisite

Must be enrolled in Graduate Diploma of Medical Biostatistics or Master of Medical Statistics to enrol in this course. Pre-requisites: must have successfully completed BIOS6020, BIOS6040, BIOS6050, BIOS6070, BIOS6170, and EPID6420.

Assessment Items

Essay: Essay