Advanced/AI-supported Rating Models for P2P systems
At a glance
- Project leader : Dr. Branka Hadji Misheva
- Co-project leader : Prof. Jörg Osterrieder, Dr. Jan-Alexander Posth, Christoph Schmidhuber
- Project team : Florian Bozhdaraj, Dr. Sandra Stupar
- Project status : completed
- Funding partner : Innosuisse (Innovationsscheck / Projekt Nr. 44596.1 INNO-SBM)
- Project partner : i2 operations GmbH
- Contact person : Branka Hadji Misheva
Description
Our innovative idea is to provide a Proof of Concept concerning
a superior and adaptive rating system applicable to both SME and
personal lending, able to cope with low-quality, sparse data that
is commonplace in the context of P2P lenders. If successful, we
will have both a new methodology (how to deal with low-quality,
sparse data in the context of credit risk) as well as a new product
that will complement our existing business.
We further intend to use our unique and comprehensive loan data to
provide transparency for other market participants and the
financial regulators. We will achieve this by providing insights on
default and migration probabilities similar to what rating agencies
(Standard&Poors, Moodys) do for traditional credit markets. An
enhanced methodology will lead to a sophisticated rating system on
the level of the loan originator as well as on the level of the
borrower.