Institutional Work

Financial Data Infrastructure

My institutional work focuses on data management. The data management includes producing, enriching, and visualizing new financial data. The Financial Data Infrastructure provides users with those financial data and additional services to boost financial research.

Description

The project „Financial Data Infrastructure“ aims to solve two fundamental problems of empirical research in the financial sector. First, the lack of research-ready pan-European financial data sets implies that many financial researchers analyze US data sets instead. The findings of their studies are therefore not easily transferable to the European financial markets. Second, empirical financial researchers often use proprietary data sets, and their incentives to share such data sets with other researchers are low, which weakens empirical research in the financial sector. Therefore, our team builds the Financial Data Infrastructure. We generate new research-ready European financial data sets and enrich them to put them into the context of research. We focus on innovating on our platform to create incentives for our users to share research data or metadata to fulfill the FAIR principles (Findable, Accessible, Interoperable, Reusable) best.

  • Data Production
  • Data Enrichment
  • Data Visualization
  • FAIR Principles

Description of ETL Process

We implement Extract-Transform-Load (ETL) processes to automatically collect, process, publish and update research data. Our flexible ETL processes cope with unstable links and various data formats and standardize research data to provide the best access to our users.

Description of Data Visualization

Our users can put research data and articlesinto context using our network visualization, displaying how articles and data are connected. They can also use this visualization to search for research data and articles via adjusting the metadata used to build the network and the number of nodes (Co-occurrences). Finally, they can download all metadata of the research data and articles in the network to get a fast overview of the essential articles in a specific research area.

Description of Data Visualisation

Our users can put research data and papers into context using our network visualization, which displays how papers and data are connected. They can also use this visualization to search for research data and papers via adjusting the metadata, which are used to build the network, and the number of nodes (Co-occurences). Finally, they can download all metadata of the research data and papers in the network at once to get a fast overview of the most important papers in a specific research area.

Data Management

Data Production

We implement ETL processes to collect, archive, and provide research-ready financial European data sets automatically .

Securities Lending

We produce data of the public sector purchase program (PSPP) and the corporate sector purchase programme (CSPP).

Eurex Repo

We produce data on the Baskets of the GC Pooling Market.

European Historical Data

We provide data on the historical money and capital database.

Data Enrichment

To put data into context, we enrich them via metadata to make them more findable and accessible.

Basic Metadata

We provide a standard documentation DOI, including a description of data, authors, date, etc.

Advanced Metadata

We generate metadata for research data based on the research papers, which use such data, including keywords, classifications, and references.

Innovative Metadata

We implement machine learning methods to extract latent research data topics to make them more comprehensive and understandable.

Data Visibility

To improve the users‘ experience on our platform, we visualize all our data and enable users download their most important findings easily.

Network Visualization

Our metadata connect all research data and papers and thus put them into context. Users can adjust the visible network via their selection of metadata.

Efficient Searching

Once our users generate their optimal network around a research data set or paper, they can quickly download the most relevant connected research data or papers.

Recommender Tool

We develop a recommender tool that  provides relevant papers and potential journals based on a article’s text and metadata.

Contact

Lennart Kraft

lennart.kraft[at]wiwi.uni[hyphen]frankfurt.de

Address

Raum 1.202 im RuW-Gebäude
Theodor-W.-Adorno-Platz 4
D-60323 Frankfurt am Main