Handbook

This course in statistics covers the design and analysis of experiments and observational studies to answer causal questions. Some examples of causal questions are: What is the reduction in mortality in cancer patients when given a new cancer drug, compared to the current standard of care? Did an increase in the minimum wage result in a change in the unemployment rate? Which website design maximises the number of users who sign up for an account on my website?

Topics in the course include potential outcomes, causal estimands, and the importance of the design phase in both experiments and observational studies. Experimental design topics will include randomisation techniques, such as blocked and factorial experiments, and analysis methods, including both parametric and randomisation methods. The experimental design material relates to fields including agriculture, natural sciences, and online experiments. Causal inference topics will include propensity scores, weighting, and matching. The causal inference material relates to fields including social sciences and public health. Students will put methods into practice using the statistical software R throughout the course.

The intended audience of this course is wide-ranging, from statistics and data science students to students in the social and natural sciences (biology, psychology, economics, and others). Prior knowledge of probability and statistical inference is necessary (prerequisite: MATH2801 or MATH2901).

This course is jointly taught at two levels. MATH3852 is for 3rd year undergraduates, whereas MATH5852 is for Honours & Masters students.