Postdoc Researcher for the project AI4EFin

At the Bucharest University of Economic Studies, the position of a Postdoc Researcher with 50% of the regular working time is to be filled as soon as possible, for the project  AI4EFin, principal investigator Prof. Dr. Stefan Lessmann. The position is limited to 3 years.

Applicants with a mathematical-quantitative profile are particularly welcome, even without a direct connection to banking and finance, if they are interested. In principle, you should have a university degree at PhD level in the field of economics, (business) mathematics, (business) informatics, statistics or similar with above-average success. Creativity, willingness to learn, scientific-oriented thinking as well as high communication and team skills should be a matter of course.

As a member of our team, you will deal with challenging questions of energy finance. Within the framework of your assignment, you will have the opportunity to present your results at international conferences. Our team offers flexible working hours and intensive cooperation in a committed team.

The application deadline is December 5, 2023. If you have any questions, please contact Prof. Daniel Traian Pele (danpele@ase.ro). You can find more details below, as well a short presentation of the project.

AI4EFin - presentation

Energy finance highlights the interdependency of energy and financial markets. While the traditional viewpoint of energy markets being a source for shocks in financial markets remains valid, the increasing financialization of energy products renders the linkage between those markets far more complex. Understanding these relationships and answering the crucial question of how to fuel world economies hunger for energy while decreasing greenhouse gas emission requires a new family of tools that turn the vast amounts of data in the energy finance ecosystem into insights for decision-making and ultimately enhance the efficiency, resilience, and sustainability of energy operations and their financing.

The initiative AI for Energy Finance (AI4EFin) speaks to these challenges. Built around a methodological core, we craft novel machine learning (ML) and artificial intelligence (AI) instruments for pattern extraction, explanation, and forecasting of the high-dimensional, non-stationary, temporal data encountered in energy finance.

We design this new family of ML/AI instruments to provide distinct features that support decision analysis and risk management in energy finance. These features include probabilistic models, which estimate the full conditional distribution of energy derivative prices and other targets. Distributional forecasts facilitate the applicability of risk management tools such as (conditional) value-at-risk and, thus, effectively support the quantification and management of financial and energy risks.

Drawing on the potential outcome framework, recent work on transfer learning in transformer networks, we also devise ML/AI instruments that model the causal effect of interventions/shocks on price developments and market outcomes. Beyond their merit for risk management, these new causal approaches also guide policymakers in devising/revising regulatory programs and other market interventions, and facilitate estimating the effectiveness of these interventions.