161704

# Bayesian Statistics

Introduction to the Bayesian paradigm. Markov Chain Monte Carlo estimation using WinBUGS. Comparison with frequentist statistics. Noninformative and improper priors. Inference and model selection. Linear and generalized linear models. Hierarchical Bayes.

## Course code

Qualifications are made up of courses. Some universities call these papers. Each course is numbered using six digits.

161704

## Level

The fourth number of the course code shows the level of the course. For example, in course 219206, the fourth number is a 2, so it is a 200-level course (usually studied in the second year of full-time study).

700-level

## Credits

Each course is worth a number of credits. You combine courses (credits) to meet the total number of credits needed for your qualification.

15

Statistics

## Course planning information

### Course notes

Access to a Windows PC and an approved statistics package is required for analysis of data.

### General progression requirements

You may enrol in a postgraduate course (that is a 700-, 800- or 900-level course) if you meet the prerequisites for that course and have been admitted to a qualification which lists the course in its schedule.

## Learning outcomes

What you will learn. Knowledge, skills and attitudes you’ll be able to show as a result of successfully finishing this course.

• 1 Describe the fundamental differences between the Bayesian and frequentist approaches to statistics.
• 2 Construct conjugate families of prior distributions for common sampling distributions.
• 3 Explain the rationale for noninformative prior distributions, and the limitations involved in using them.
• 4 Carry out Bayesian inference procedures such as point and interval estimation and hypothesis testing, for common sampling distributions.
• 5 Construct, in a Bayesian framework, hierarchical versions of common statistical models including Generalised Linear Models.
• 6 Use suitable software tools to set up and analyse Bayesian statistical models using Markov chain Monte Carlo (MCMC) methods.

Learning outcomes can change before the start of the semester you are studying the course in.

## Assessments

Assessment Learning outcomes assessed Weighting
Written Assignment 1 2 3 4 6 30%
Written Assignment 4 5 6 30%
Written Assignment 1 2 3 4 5 6 40%

Assessment weightings can change up to the start of the semester the course is delivered in.

You may need to take more assessments depending on where, how, and when you choose to take this course.

### Explanation of assessment types

Computer programmes
Computer animation and screening, design, programming, models and other computer work.
Creative compositions
Animations, films, models, textiles, websites, and other compositions.
Exam College or GRS-based (not centrally scheduled)
An exam scheduled by a college or the Graduate Research School (GRS). The exam could be online, oral, field, practical skills, written exams or another format.
Exam (centrally scheduled)
An exam scheduled by Assessment Services (centrally) – you’ll usually be told when and where the exam is through the student portal.
Oral or performance or presentation
Debates, demonstrations, exhibitions, interviews, oral proposals, role play, speech and other performances or presentations.
Participation
You may be assessed on your participation in activities such as online fora, laboratories, debates, tutorials, exercises, seminars, and so on.
Portfolio
Creative, learning, online, narrative, photographic, written, and other portfolios.
Practical or placement
Field trips, field work, placements, seminars, workshops, voluntary work, and other activities.
Simulation
Technology-based or experience-based simulations.
Test
Laboratory, online, multi-choice, short answer, spoken, and other tests – arranged by the school.
Written assignment
Essays, group or individual projects, proposals, reports, reviews, writing exercises, and other written assignments.

## Textbooks needed

Textbooks can change. We recommend you wait until at least seven weeks before the semester starts to buy your textbooks.

### Compulsory

#### BAYESIAN IDEAS AND DATA ANALYSIS: AN INTRODUCTION FOR SCIENTISTS AND STATISTICIANS.

Author
CHRISTENSEN R, JOHNSON W, BRANSCUM A, HANSON TE
Edition
2011
Publisher
CRC Press

Campus Books stock textbooks and legislation. For more information visit Campus Books.

## No offerings available

There are currently no offerings available for this course. Search for a different course.