620-371 Linear Models

Passing 620-371 precludes subsequent credit for 620-270 or 620-272.

One of 620-202, 620-270 or 620-272.

This subject develops problem-solving skills and sharpens analytical skills. Students will work in groups, tackling unfamiliar problems. Each team will plan their project work and deliver oral and written presentations.

This subject introduces the basic theory of the general linear model and explains how linear models are used to analyse data. Students should develop the ability to examine data for common structures and patterns and to formulate linear models in specific practical situations, including univariate normal responses with a combination of explanatory factors. They learn to carry out the necessary computations on the computer, check the assumptions of the model in specific situations, and express the results of modelling in scientifically useful terms. This subject demonstrates the importance of the general linear model in analysing a variety of data and giving useful information about scientific subject matter.

Topics covered include general least squares theory of estimation and hypothesis testing; application to one and two-way classifications; factorial experiments; analysis of covariance; multiple regression; polynomial regression; use of statistical computer packages; nested and crossed factors; fixed and random effects; and multiple and orthogonal contrasts.

Up to 50 pages of written assignments due during the semester (20%); a group project during semester with a 30-minute oral presentation and a written report of up to 30 pages (15%); a 3-hour written examination in the examination period (65%).