Page 174 - FINAL_HANDBOOK_20252026
P. 174
Faculty of Science Handbook, Academic Session 2025/2026
Multivariate control charts. Acceptance methods: principal component analysis,
sampling plans. and linear discriminant analysis. Clustering
methods for unsupervised learning.
Assessment: Application of linear discriminant analysis,
Continuous Assessment: 40% classification and regression trees for
Summative Assessment: 60% supervised learning.
Assessment:
SIT3012 Continuous Assessment: 40%
DESIGN AND ANALYSIS OF EXPERIMENTS Summative Assessment: 60%
Philosophy related to statistical designed
experiments. Completely randomized one- SIT3016
factor design. Randomized block designs. GENERALIZED LINEAR MODELS
Latin squares. Incomplete block designs.
Factorial designs. Confounding. Fractional Introduction to generalized linear model
factorial designs. based on the exponential family. For
example, multiple linear regression for
Assessment: normal data, logistic regression for binary
Continuous Assessment: 40% data, Poisson regression for counts, log
Summative Assessment: 60% linear for contingency table, and gamma
regression for continuous non-normal data.
Study the theory of GLM including
SIT3013 estimation and inference.
ANALYSIS OF FAILURE AND SURVIVAL
DATA Introduction to fitting GLM in R.
Survival distributions, hazard models. Focus on the analysis of data: binary, count
Reliability of systems, stochastic models. and continuous, model selection, model
Censoring and life-tables. The product-limit evaluation, interpretation, prediction and
estimator. Parametric survival models residual analysis.
under censoring. Cox proportional hazards
model and other models with covariates. Assessment:
Continuous Assessment: 40%
Assessment: Summative Assessment: 60%
Continuous Assessment: 40%
Summative Assessment: 60% SIT3017
STATISTICAL LEARNING AND DATA
SIT3015 MINING
INTRODUCTION TO MULTIVARIATE
ANALYSIS This course prepares students for applied
work in data science by building on
Matrix algebra and random vectors. students’ foundations of data science skills.
Multivariate normal distribution. Wishart Students will learn advanced methods in
distribution and Hotelling distribution. statistical learning and data mining, using
Multivariate linear regression, canonical appropriate computing tools such as R. The
correlation analysis. Dimensional reduction strengths of the diversity of approaches are
175
