Towards a European financial data space

Work Package 1
Towards a European financial data space.

Leading institution

Useful links

Overview

Data is in the centre of the digital transformation. By ensuring that all dimensions (accuracy, consistency, completeness, currency, volatility and timeliness) of data quality are satisfied, that we have sufficiently large, high-quality data available and that we can detect dependencies in high dimensions, high frequency and high veracity financial data, DIGITAL is in a position to move towards a European financial data space and make use of substantially more data sources than today.

Under this research topic, an unprecedented collection of innovative data quality methodologies and data augmentation techniques will be applied to a wide variety of industry-relevant datasets. Overcoming the obstacles of data quality and availability (through novel or extended large language models, deep generation of data, and anomaly- and dependence detection models baked on network concepts) will contribute to the literature and further provide a valuable tool for the European Finance industry to enhance product offerings, reduce financial market risks, and work toward a European financial data space. With quality data available and the ability to detect dependencies in high dimensions, high frequency, and high veracity financial data, DIGITAL is in a position to move towards a European financial data space and make use of substantially more data sources than today.

WP 1 Team

Codruta Mare

WP1 Lead

Research topics

Under this research stream, FOUR doctoral candidates will tackle the following research projects:

Collaborative learning across data silos

Description: Connecting several dozen different data pipeline components and integrating an excessive number of APIs to leverage siloed data is a significant barrier to the comprehensive implementation of AI-based systems in finance. Currently, very little research is devoted to addressing all of the challenges associated with training, testing, and deploying cutting-edge ML and DL methods while leveraging siloed data. We will concentrate on the data challenges that finance service providers face by proposing solutions to streamline data collection, resolve data quality issues, and structure data to support downstream processes.

Host: WU Vienna University of Economics and Business

Planned secondments:

Swedbank AB (SWE), Prof. Dr. Tadas Gudaitis, M23, 18 months, research on prototype implementations, applied research

Fraunhofer Institute (FRA). Prof. Dr. Ralf Korn, M41, 4 months, applied industry-research, exposure to world-leading research centre and infrastructure

Detecting anomalies and dependence structures in high dimensional, high frequency financial data

Description: Herding, a well-known financial anomaly, is thought to cause high volatility, volatile prices, and low liquidity (Bikhchandani and Sharma, 2000). Greed and herd behaviour caused the seventeenth-century tulip mania, the 1995–2000 Internet bubble, and the 2015 Chinese stock market crash. This project studies high-dimensional sentiment networks and herd behaviour on the stock market. To better fit investor sentiment, the project will calibrate the option pricing model, Stochastic Volatility and Correlated Jump (SVCJ).

Host: Bucharest University of Economic Studies, Romania (ASE)

Planned Secondments:

DeutscheBank (DBA). Roman Timofeev, M27, 6 months, contribute datasets, expertise on applications of AI and anomaly detection and early warning systems, as well as expertise on predictive analytics, semantic analysis and risk management.

Royalton (ROY). Dr. Michael Althof, M33, 12 months, for training in portfolio optimization of ETFs

Predicting financial trends using text mining and NLP

Description: The main goal of this doctoral candidate is to enhance the application of AI-driven natural language processing (NLP) techniques for forecasting credit risk and detecting financial fraud. This involves analyzing speech text from audit reports, social media, and various other sources. The focus is on predicting non-compliance by interpreting the sentiment in free-text answers from survey participants. Additionally, the project aims to develop attitudinal indices derived from free text and integrate them into behavioral models. These models, which also consider other qualitative and quantitative elements, will help in assessing the probability of system fraud and determining the risk level associated with accreditation processes.

Host: The Babes-Bolyai University

Planned Secondments:

Raiffeisen Bank International (RAI), Dr. Stefan Theußl, M15, 18 months, research exposure in a global business environment, trend modelling

European Central Bank (ECB), Dr. Lukasz Kubicki, M33, 12 months, exposure to globally leading central bank, research training on EU principles, supervision

Deep Generation of Financial Time Series

Description: Macroeconomics factors such as central banks’ interest rates, inflation, unemployment rate, house price indices, to name a few, are of foremost importance in Financial Markets. The aim of this project is to benchmark various methods from classical statistical learning and modern machine learning in order to predict their point value in the future. As a second step the student will be using the above predictions to forecast future market scenarios in a what-if fashion.

Host: Cardo S.R.L

Planned Secondments:

WU Vienna University of Economics and Business (WWU), Prof. Dr. Kurt Hornik, M27, 14 months, theoretical modelling and mathematics for deep learning

Fraunhofer Institute (FRA), Prof. Dr. Ralf Korn, M41, 4 months, applied industry-research, exposure to world-leading research centre and infrastructure