Thesis Project

Raphaela Kotsch

"The thesis aims to develop machine learning algorithms that (i) explain trading patterns of market participants, (ii) detect and predict fraudulent behavior and (iii) classify and quantify countries’ positions on carbon markets under the Paris Agreement on the basis of texts submitted to the UNFCCC negotiations."

Homepage

Thesis: Applying machine learning to carbon markets: Building trust for a net-zero emissions world

Supervisors:
Prof. Dr. Katharina Michaelowa (UZH)
Prof. Dr. Regina Betz (ZHAW)

Keywords: carbon pricing, emissions trading, machine learning, fraud detection, market behavior

Abstract: The PhD thesis applies machine learning to empirically examine challenges in the design, the implementation and the monitoring of carbon markets. The topic of the thesis is thus located at the interface between data science, social science and climate policy. Carbon markets are an economic instrument that is used at the international, as well as the national and regional level to reduce greenhouse gas emissions and to ultimately tackle climate change. Today, there are over 15 carbon markets globally such as in Switzerland, the European Union and California and the number is expected to grow with around 60 initiatives of jurisdictions planning to implement a carbon price as part of their commitments under the Paris Agreement. Yet, our understanding of the functioning of these markets and the behavior of markets participants is limited. The thesis aims to develop machine learning algorithms that (i) explain trading patterns of market participants, (ii) detect and predict fraudulent behavior and (iii) classify and quantify countries’ positions on carbon markets under the Paris Agreement on the basis of texts submitted to the UNFCCC negotiations.

Planned Duration: 2019-2022