Kamakura upgrades credit default prediction software

The company said version 3.0 of its Jarrow-Chava default probability KDP-jc is an improvement on previous models that tended to overestimate the number of defaults.

The software uses ‘hazard rate’ modelling to predict the number of public company defaults, combining macro-economic factors, financial ratios and stock price and index information.

Leonard Matz, managing director of marketing at Kamakura, said the model, completed this month, would be automatically available to clients with older versions. He declined to specify names, but said there were “about a dozen” customers. Matz also declined to comment on cost.

Donald van Deventer, chief executive of Kamakura, called the model “a significant step forward” in addressing the problems of previous models that had “an excess of false predictions of default, a modest degree of accuracy and a surprisingly low correlation between the actual and expected number of defaults". Van Deventer added that the model had been tested for three dimensions of risk: ordinal ranking of companies by riskiness, consistency between actual and predicted defaults, and minimisation of false positives. In all tests the new version has outperformed older models, said Kamakura.


Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

Calibrating interest rate curves for a new era

Dmitry Pugachevsky, director of research at Quantifi, explores why building an accurate and robust interest rate curve has considerable implications for a broad range of financial operations – from setting benchmark rates to managing risk – and hinges on…

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here