University of Auckland 2016, 2017 & 2023—present

2024

Key service roles:

Postgraduate Adviser (Data Science)
STATS 101/101G/108 Technical Lead
Undergraduate Staff Student Consultative Committee Meeting Chair

Southwest University, Chongqing, China - teaching partnership (Semester 1)

A short course as part of the partnership between SWU and UoA. Mid-May to early June.

STATS 101/108 (Semester 2)

Intended for anyone who will ever have to collect or make sense of data, either in their career or private life. Steps involved in conducting a statistical investigation are studied with the main emphasis being on data analysis and the background concepts necessary for successfully analysing data, extrapolating from patterns in data to more generally applicable conclusions and communicating results to others. Other topics include probability; confidence intervals, statistical significance, t-tests, and p-values; nonparametric methods; one-way analysis of variance, simple linear regression, correlation, tables of counts and the chi-square test.

More information on the course is available in the Course outline. There are multiple course codes as this is an important service course for Science, Arts and Business students.

STATS 330 (Semester 2)

R is the programming language used. Familiarity from STATS 201/208 is expected.
I use and teach a lot of tidyverse, especially dplyr for data wrangling and ggplot for visualization.
- If you’re not familiar with these, consider checking out:
  - chapters 3 and 5 of R for Data Science by Hadley Wickham & Garret Grolemund, and
  - the DoSS Toolkit, especially the modules ‘Hand me my plyrs’ and To ggplot or not to ggplot.

STATS 330 further develops ideas introduced in STATS 201/208, giving a synthesis and broader understanding of generalised linear models and related methods. Simulation-based procedures, including bootstrapping and cross-validation, are introduced as a means to provide robust inference, investigate the consequences of assumption violations, and solve goodness-of-fit and model-selection problems. Particular focus is placed on how the modelling procedure varies depending on whether the analysis aims to explain an underlying process or predict future observations. Students will learn to implement all methods taught in R, the widely used, open-source software environment for statistical computing. Emphasis is on practical application, providing students with a versatile statistical toolbox useful for a range of fields in both academia and industry, including almost all subjects in Business and Economics, along with any experimental or social science. It is also a useful complement to Computer Science.

~ Course description from the course outline

2023

STATS 101/108 (1st half of S2)

This course has recently undergone a significant rewrite by the STATS 10x team including Anna Fergusson, Thomas Lumley, Emma Lehrke, Lars Thomsen and others. It has a focus on interactive lectures and modern data considerations.

I was of the teaching team in second semester.

More information on the course is available in the Course outline. There are multiple course codes as this is an important service course for Science, Arts and Business students.

STATS 369 (1st half of S2)

This course has been developed by Thomas Lumley (Department of Statistics), with additional teaching and development by Lisa Chen (Statistics), David Welch (Computer Science), Yalu Wen (Statistics).

This is a course on predictive modelling using real data. STATS 369 is required for the major in Data Science but is also taken by undergraduate students from other majors in Science, Engineering, and Commerce who are interested in careers involving large-scale data analysis and modelling. This course is good preparation for anyone wanting to do postgraduate study in Data Science. We emphasise understanding the modelling techniques in addition to being able to apply them using R. The predictive techniques covered include linear regression and discrimination, tree-based models, and neural networks. The course also covers the data cleaning and manipulation needed to prepare real-world data for analysis and some of the ethical issues that arise from the use of automated predictive models. The skills developed in this course are particularly useful for those wishing to have a career involving data science and predictive modelling, which are areas in high demand.

~ Course description from the Course outline

STATS 150/150G (1st 5 classes of S2)

This course is coordinated by Stephanie Budgett. I taught the first 5 classes on media reports and surveys and polls.

STATS 150 aims to prepare anyone, regardless of whether or not they have any background in statistics, to become a critical consumer of statistical information. STATS 150 will be useful for aspiring journalists, politicians, political scientists, sociologists, lawyers, public communicators, health personnel, conservationists, environmental scientists, business people, marketers, engineers, and scientists. It examines the uses, limitations, and abuses of statistical information in a variety of activities such as polling, public health, sport, law, marketing, and the environment. The statistical concepts and thinking underlying data-based arguments will be explored. The interpretation and critical evaluation of statistically-based reports as well as the construction of statistically sound arguments and reports will be emphasised. Some course material will be drawn from topics currently in the news.

~ Course description from the the Course outline

STATS 330 (2nd half of S1)

Co-teaching one stream of 100 students with Prof. Alain Vandal.

Course information.

R is the programming language used in my courses.
I tend to use and teach a lot of tidyverse, especially dplyr for data wrangling and ggplot for visualization.
- If you’re not familiar with these, consider checking out:
  - chapters 3 and 5 of R for Data Science by Hadley Wickham & Garret Grolemund, and
  - the DoSS Toolkit, especially the modules ‘Hand me my plyrs’ and To ggplot or not to ggplot.

2017

In 2017 I was a Professional Teaching Fellow for 6 months, teaching STATS101/101/108: Introduction to Statistics. I was awarded a Department of Statistics Teaching Award (Guest) for my work over this period.

Univeristy of Toronto 2020-2023

2022–23

I wil be teaching the Winter semester of STA490. Lectures for this course will be delivered online.

R is the programming language used in my courses.
I tend to use and teach a lot of tidyverse, especially dplyr for data wrangling and ggplot for visualization.
- If you’re not familiar with these, consider checking out:
  - chapters 3 and 5 of R for Data Science by Hadley Wickham & Garret Grolemund, and
  - the DoSS Toolkit, especially the modules ‘Hand me my plyrs’ and To ggplot or not to ggplot.

While I will not be supervising any reading courses, I highly recommend strong final year students look into them. STA496/497: Readings in Statistics must be registered for as part of a special enrolment that usually happens over the summer.