PP433      Half Unit
Topics in Model Based Quantitative Analysis for Public Policy

This information is for the 2025/26 session.

Course Convenor

Availability

Requisites

Pre-requisites:

Students must have completed PP455 before taking this course.

Additional requisites:

Requires successful completion or exemption from PP455 or an equivalent course.

Basic knowledge in R or an equivalent programming language is required. Students who do not have prior knowledge of R will be required to take an R module with the Digital Skills Lab.

Data often are structured in a manner that we can explicitly model, such as the correlation of a macroeconomic variable (e.g. inflation) measured at different points in time or correlations amongst students within a classroom and multiple classrooms nested within a school. Without modelling these forms of dependence, our estimates may be biased and less precise than traditional inference techniques would report.

This course provides a hands-on introduction to model-based inference strategies including prediction, forecasting models and applications of simulation analysis and Bayesian models. The course begins by distinguishing forecasting and prediction accuracy as distinct goals in statistical learning. Initial weeks focus on multilevel modelling and random effects and contrasts these approaches from fixed effects analysis. Next the course introduces modelling techniques and diagnostic tools for univariate time series analysis and forecasting. Key concepts include stationarity, autocorrelation and partial autocorrelation. Time series models begin with basic autoregressive and moving average models and build up to ARIMA / SARIMA models for seasonal data. If time permits, the final weeks of the course will introduce Bayesian analysis and regression and discuss the role of defining and incorporating prior knowledge and expectations into statistical analysis along with a motivation of how MCMC techniques can be used to obtain estimates from these models. Lessons will include analysis and examples from education, international finance, urban policy and healthcare.

15 hours of seminars and 20 hours of lectures in the Winter Term.

This course has a reading week in Week 6 of Winter Term.

Student groups will be expected to produce one piece of formative work in the WT associated with the final summative assessment.

• Gelman, Andrew, Jennifer Hill, and Aki Vehtari. Regression and other stories. Cambridge University Press, 2021.
• Kruschke, John. "Doing Bayesian data analysis: A tutorial with R, JAGS, and Stan." (2014).
• Hyndman, Rob J., and George Athanasopoulos. Forecasting: principles and practice. OTexts, 2018.

Project (75%)

This component of assessment includes an element of group work.

Report (25%)

The final assessment consists of a group project (75%) where groups identify an existing piece of published work and replicate the core results the original analysis. Groups will then motivate and apply techniques from the course to expand upon the chosen paper’s analysis.