Kamakura Corporation Announces Non-Public Firm Default Model

Model is Seamlessly Integrated with KRIS and KRM.

Honolulu-based Kamakura Corporation announced Wednesday that the Kamakura Risk Information Services Non-Public Firm Default Model is immediately available on the KRIS web site and within Kamakura Risk Manager. The Kamakura Non-Public Firm Model is based on 2.866 million annual observations of non-public firms and 41,119 company failures in a multi-national non-public firm data base. The KRIS Non-Public Firm Model uses six key inputs from Kamakura’s industry leading KRIS Public Firm Models, first launched in 2002. The KRIS Non-Public Firm Model displays default probabilities on an annualized, cumulative and forward basis for a 1, 2 and 3 year term.

“In the wake of the 2007-2009 credit crisis and the well-known performance of legacy rating agencies, Kamakura’s risk information and risk management software clients have strongly encouraged Kamakura to launch a state of the art non-public firm model,” said Kamakura Chief Administrative Officer Martin Zorn, “As a banker with three decades of lending experience, I can say that the new KRIS Non-Public Firm Model brings credit risk insights that are orders of magnitude better than existing models from both an accuracy and a stress testing perspective. We look forward to working with Kamakura clients around the world to help them immediately profit from these insights.”

The Kamakura Non-Public Firm Model subscription includes a complete technical guide authored by Professor Robert A. Jarrow, Kamakura’s Managing Director for research, and other Kamakura colleagues. The technical guide includes all inputs and all coefficients used in the model and a complete suite of Basel-focused accuracy tests. Because of the complete integration of KRIS version 5.0 public firm default probabilities in the KRIS Non-Public Firm Model, the long term accuracy of the Non-Public Firm Model is actually even better than older versions of KRIS public firm models. Those older public firm models, in turn, were more accurate than legacy ratings on public firms.

• Read more under:
• BankingTech
• InsurTech