ST4233 Linear Models

Module Description

Linear statistical models are used to study the way a response variable depends on an unknown, linear combination of explanatory and/or classification variables. This module focuses on the theory of linear models and possible topics include: linear regression model, general linear model, prediction problems, sensitivity analysis, analysis of incomplete data, robust regression, multiple comparisons, introduction to generalised linear models, nonlinear regression. This module is targeted at students who are interested in Statistics and are able to meet the prerequisite.

Lecture Topics
- Matrix and Vector
- Random Vector and Matrix
- Multivariate Normal Distribution
- DF of Quadratic Forms in y
- Simple Linear Regression
- Multiple Regression: Estimation
- Multiple Regression

Core module for statistics major.

Workload
1.5 hours lecture every week
Lectures were taught by Prof Zhou Wang.

One hour tutorial
Tutorials were held after the lecture.

Lectures were not webcasted.

Assessment (AY11/12 Sem 2)
30% Assignment
70% Finals

The assignment consisted of 7-8 questions to be completed in 2 weeks.

Two pieces of cheatsheet is allowed for the final paper. Most of the chapters were tested.

Personal
Waiting for results

This module is like a continuation of ST3131 except that we will learn to handle everything in matrix. 2 out of 6 questions were on proofs.