Registration link: https://forms.gle/TXVGX2SdvzuAB8LR6

Registration is free (lunch + dinner included)

Spaces are limited, so please register early to secure your place.

The Bucharest University of Economic Studies, through the IDA — Institute for Digital Assets and the AI4EFin — Artificial Intelligence for Energy Finance Research Group, is proud to host the AIDA Conference — AI & Digital Assets in Energy and Finance.

This conference is dedicated to advancing the rigorous statistical and theoretical understanding of two rapidly evolving fields: digital assets and energy finance. It aims to convene leading academics to address critical open questions and deepen our analytical frameworks for these transformative domains in finance and beyond.

The conference will explore key directions and foster discussions on the following pivotal areas:

How do we formally characterize and model the discrete, often indivisible, units of digital assets? What are the implications of this granularity for market microstructure, price formation, liquidity, and order book dynamics, especially in the absence of traditional market-making structures?

What novel statistical models are required to capture the heavy-tailed distributions, jumps, and regime-switching behaviors observed in digital asset prices? How can we effectively model and forecast their often extreme volatility, considering the unique informational and transactional characteristics of blockchain data?

How do the underlying distributed ledger technologies and network effects influence the statistical properties of digital assets? What methodologies can be employed to quantify and model the complex interdependencies between different digital assets, protocols, and the broader financial ecosystem?

Given the distinct statistical properties and operational risks associated with digital assets, what robust statistical frameworks are necessary for effective risk measurement, stress testing, and portfolio optimization? How can traditional risk models be adapted — or new ones developed — to account for the unique characteristics of this asset class?

How can statistical inference be applied to understand the causal impact of various protocols, governance mechanisms, and economic incentives within the DeFi landscape? What statistical tools are appropriate for analyzing and designing efficient and robust mechanisms in decentralized environments?

How can artificial intelligence and machine learning methods improve the modelling, forecasting, and risk management of energy markets? What novel approaches — including neural forecasting, large language models, and multimodal AI — can capture the complex dynamics of electricity prices, carbon markets, and renewable energy integration? How do digital assets and blockchain-based mechanisms intersect with energy trading, green finance, and the transition to sustainable energy systems?

How can advanced machine learning and deep learning architectures improve short-term and long-term electricity price forecasting? What role do neural forecasting models, transformer architectures, and hybrid approaches play in capturing the complex seasonality, spikes, and negative prices characteristic of modern electricity markets?

How can AI methods address the challenges of integrating intermittent renewable energy sources into power grids? What novel forecasting and optimization techniques can improve the management of solar, wind, and storage assets, and how do these interact with energy trading strategies?

How can blockchain technology and tokenization transform energy trading, peer-to-peer energy markets, and renewable energy certificate systems? What are the implications of decentralized energy markets for market efficiency, transparency, and the energy transition?

This conference seeks contributions that push the boundaries of current statistical and theoretical methodologies. We invite researchers to engage in discussions that will lay the groundwork for a more profound and statistically grounded understanding of these transformative fields.

Matthias R. Fengler

Google scholar profile

From Digital Transactions to Real Spending: Insights from a Novel Consumption Index for Switzerland

Matthias Fengler is Professor of Econometrics at the School of Economics and Political Science at the University of St. Gallen (HSG). He studied at the Eberhard Karls University in Tübingen, the University of Mannheim, UC Berkeley, and the Humboldt-University in Berlin. His research focuses on asset pricing, volatility modeling, risk-management, and the analysis of financial text data. Together with colleagues from academia and industry, he runs the project Monitoring Consumption Switzerland and has recently launched the Consumer Spending Index (CSI) for Switzerland.

Rafał Weron

Google scholar profile

Machine learning for electricity price forecasting

Rafał Weron is Professor of Economic Sciences at the Wrocław University of Science and Technology. He is a member of the Center for Research in Energy (CoRE) at Aarhus BSS, Denmark, and the Academia Europaea. With a Ph.D. (1999) in Financial Mathematics and a habilitation (2009) and professor title (2015) in Economic Sciences, he is one of the leading world experts on energy forecasting. His other research interests include computational economics, computational statistics and machine learning, financial engineering and risk management. He is the author of five books and 100+ peer-reviewed book chapters and journal articles, including Hot and Highly Cited Papers.

Programme: TBA

Useful links: