Bayesian Scenario-Based Projection of Climate-Related Changes in Global Economic Growth

Overview

This interdisciplinary project investigates how Bayesian statistical modelling can be used to quantify uncertainty in projected climate-related changes in global economic growth under alternative future climate scenarios. The project will combine historical country-level economic growth data with climate variables such as temperature and precipitation, estimate a Bayesian climate-response model, and use simulated future climate variables from different scenario pathways to produce probabilistic projections of future economic outcomes.

The project connects statistics, econometrics, climate science, data science, and global economic risk analysis. It is an exploratory and methodological project.

This topic is not currently attached to an existing externally funded project.

Research engagement

The student will engage with literature on climate-economy modelling, Bayesian statistics, scenario-based projections, and uncertainty quantification. They will also work with real-world economic and climate data, explore how climate variables can be linked to economic outcomes, and consider how modelling assumptions affect projected results.

Research activities

The student will work with the supervisor to:

- review relevant literature on climate change and economic growth;

- identify and prepare suitable country-level GDP and climate datasets;

- conduct exploratory data analysis and visualise historical climate-economy relationships;

- implement a Bayesian regression or hierarchical model using R, Python, or Stan;

- generate scenario-based projections using future simulated climate variables;

- summarise posterior uncertainty using credible intervals and visualisations;

- prepare a short research report, presentation, or reproducible notebook summarising the project findings.

Research skills

The student will develop skills in:

• literature review and research problem formulation;
• data cleaning and reproducible data analysis;
• Bayesian inference;
• uncertainty quantification and interpretation of credible intervals;
• scenario-based projection methods;
• visualisation of statistical results;
• scientific communication and research presentation.

Outcomes

The expected outcomes of the project are:

• a cleaned and documented country-level climate-economy dataset or workflow;
• a Bayesian model for projecting climate-related changes in economic growth;
• scenario-based projection results under alternative climate pathways;
• visualisations of projected outcomes and uncertainty;
• a short report, presentation, or reproducible notebook that could support future research, student project development, or preliminary work toward a larger climate-economy modelling study.

Skills and experience

This project would suit a student with an interest in statistics, econometrics, data science, climate risk, or applied economic modelling. Some experience with regression modelling and basic programming in R or Python would be helpful. Prior experience with Bayesian modelling is desirable but not required, provided the student is willing to learn. The ideal student should be comfortable working with quantitative data and able to work independently with regular supervisory guidance.

Start date

2 November, 2026

End date

19 February, 2027

Location

QUT Gardens Point campus, with supervision meetings either in person or online as required.

Keywords

• Bayesian modelling
• climate change
• economic growth
• climate risk
• econometrics
• uncertainty quantification
• data science

Contact

Dan Li

3138 1719

d33.li@qut.edu.au