STA409 Extended Linear Models (8)

This subject extends the linear model into situations involving non independent observations. Such correlated data arise in repeated measures, spatial data and time series observations. Advanced techiques such as REML and GLMMs will be covered, but the modifying constraint will be the availability of solutions in software that will allow familiarisation via empirical demonstations on selected problems.

No offerings have been identified for this subject in 2020.

Where differences exist between the Handbook and the SAL, the SAL should be taken as containing the correct subject offering details.

Subject Information

Grading System

HD/FL

Duration

One session

School

School of Science and Technology

Assumed Knowledge
STA408

Learning Outcomes

Upon successful completion of this subject, students should:
  • Be able to recognise situations in which data dependence is a key component;
  • Be aware of the consequences of lack of independence in data;
  • Appreciate the similarities and differences between various diverse applications of dependent data such as spatial statistics, time series and repeated measures;
  • Be able to analyse basic dependent designs for repeated measures and variance components via available software;
  • Be able to test any assumptions involved in the use of the techniques;
  • Be able to report the results of analyses.

Syllabus

This subject will cover the following topics:

Overview of correlated data using general examples; Consequences of dependent errors; extensions of the independent error linear model: Regression - simple autoregressive models - time series models (time domain) - generalised least squares (GLS) Experimental Design - repeated measures - random effects model - mixed models - unbalanced designs - variance components (REML) - GLMMS; Tests of assumptions; Interpretation.

Residential School

This subject contains a 2 day Optional Residential School.

The information contained in the CSU Handbook was accurate at the date of publication: October 2020. The University reserves the right to vary the information at any time without notice.

Back