Teaching
I have been lecturing Statistics since 2003 at Stellenbosch University. On average I have lectures for about 7 hours per week, for 27 weeks in a year (189 hours per year). Since 2003 I have had the privilege of lecturing Statistics to thousands of undergraduate and postgraduate students. My proudest moments have always been to see my students graduate and become successful in their own careers. Below is a list of modules I have been teaching.
Current undergraduate modules
Statistics and Data Science 188
This is a year module taught to first year BCom, BSc and BA students at Stellenbosch University. In this module I teach exploratory data analysis, introductory probability theory, statistical inference, and regression analysis. This module also has a strong focus on applications in MS Excel.
Prescribed textbook: Levine, Stephan, Krehbiel, and Szabat, Statistics for Managers Using Microsoft Excel, 9th Edition, Pearson.
Statistics 318
This is a semester module taught to BCom and BSc students at Stellenbosch University. This module consists of two parts: Part A: Regression analysis and Part B: Multivariate methods. I teach Part B, which contains linear algebra, the multivariate normal distribution and its properties, statistical inference about mean vectors and linear discriminant models. The statistical software R and SAS are used in this module. RStudio is used as IDE for writing projects.
Statistics 348
This is also a semester module taught to BCom and BSc students at Stellenbosch University. This module also consists of two parts: Part A: Advanced statistical inference and Part B: Time series analysis, Bayesian inference and Stochastic simulation. I am responsible for teaching Part B. The Time series analysis section includes Exponential smoothing methods (Holt's linear and Holt-Winters methods), theory and applications of ARIMA models. The Bayesian inference section introduces posterior distributions, credible intervals and hypothesis tests using the normal distribution. The Simulation section includes random number generation (univariate and multivariate), Inverse Transform method, Acceptance-Rejection method, Monte Carlo integration. All applications and computations for this module are performed in R using RStudio.
Mathematical Statistics 344
This is a semester module taught to BCom, BDatSci and BEng students at Stellenbosch University. This module also consists of two parts: Part A: Stochastic processes and Part B: Statistical learning techniques. I teach Part B, which include mainly supervised learning (regression and classification). Learning techniques studied in this section: Logistic regression, Linear and Quadratic discriminant analysis, Fisher’s linear discriminant analysis, regression and classification trees, random forests, support vector classification, support vector regression, novelty detection and kernel Fisher discriminant analysis. Resampling techniques, methods for class imbalance and scaling methods are also introduced. The statistical software R is used in this module for practical applications.
Prescribed textbook: James, G., Witten, D., Hastie, T. and Tibshirani, R., (2021), An introduction to statistical learning. Springer.
Current postgraduate modules
Multivariate Statistics
This is a year module and I teach only the second semester. The first semester contains the important theoretical and mathematical aspects for Multivariate Statistics. The second semester contains the important theory and applications of the following Multivariate techniques: Inference about mean vectors (MANOVA), Regression analysis (incl. kernel ridge regression), Principal component analysis (incl. kernel PCA), Factor analysis, Discriminant analysis (incl. kernel Fisher discriminant analysis and kernel logistic regression), Clustering analysis, Multidimensional scaling, Correspondence analysis, Biplots and Procrustes analysis. The statistical software R and SAS are used extensively throughout this module, with RStudio as IDE for writing reports.
Prescribed textbook: Johnson, R.A. & Wichern, D.W. (2007), Applied multivariate statistical analysis, 6th Edition, London: Prentice-Hall international.
Past modules taught
Statistical Methods 176 and Statistics 186
These modules were taught to BCom and BAcc students (respectively) at Stellenbosch University. They merged in 2020 to become a new module known as Statistics and Data Science 188. I taught these modules for over a decade to BCom and BAcc students.
Mathematical Statistics 316 (Regression and Predictive Modelling)
Fitting regression models by means of matrices. The multiple linear regression model. Inference in the multiple linear regression model. Residual analysis. Variable selection techniques. Ridge regression. Lasso regression. Linear methods for classification. The use of R software to fit models in practice. I taught part of this module in 2020 to BCom students.
Biometry 242 (Experimental design and analysis, Department of Genetics at US)
Treatment and experimental design; efficiency of estimation; analysis of variance: F-test for homogeneity of variance, one- and two-sample hypothesis tests for means, multiple comparisons procedures; confidence intervals; non-parametric tests. All data will be analysed using applicable software. I taught this module in 2022 to BSc students.
Extramural courses taught
In this program I taught an Introductory Statistics course for grade 12 learners. This was a semester course I taught in 2007.
Maths4Stats
The Department of Statistics at UWC hosted this one-day workshop for high school teachers for a few years 2013-2015. I volunteered to teach the Introduction to probability topic.