The course is designed to emphasise the principles of statistical modelling through the cycle of proposing a model, fitting it, criticising it, and modifying it. The use of graphical methods will also be stressed.

Introduction to S-Plus. The simple linear model: statistical inference for simple linear regression, estimation and model diagnostics; transformation models; interpretation of simple linear models. Multiple regression models: matrix formulation; estimation and diagnostics; outliers and measures of influence; variable selection. Models with indicator variables (including experimental designs).

On completion of this course students should have an understanding of and be able to apply the techniques outlined in the course description.

The following final assessment scheme is proposed:

2 Assignments, 10% (5% each)

Midterm-exam, 25% or 0% (if redeemed)

Three-hour final examination, 65% or 90% (if midterm redeemed)

Students enrolled in STAT6038 will be required to prepare additional tutorial exercises, assignment exercises and examination questions whose nature and difficulty are commensurate with the graduate level nature of the course. Additional material delivered to STAT6038 students will be clearly and appropriately marked.

At least 10 hrs/week

with STAT8038 Regression Modelling

See Course Outline:  http://ecocomm.anu.edu.au/courses/outline/STAT2008.pdf

For further information please refer to http://ecocomm.anu.edu.au/courses/course.asp?code=STAT6038