Greifenberg Analytics is a state-of-the-art Fintech platform that applies Artificial Intelligence, Machine Learning, and Natural Language Processing to evaluate the risk/reward characteristics of corporate bonds. The Greifenberg platform is the only risk management system to cover the whole Chinese corporate bond market, a nearly $10 trillion market that is now the world’s largest source of fixed-income yield.
Our model uses Machine Learning to integrate a suite of risk measures, and accurately estimate default risk for individual corporate bonds as well as bond portfolios. Our inputs include:
- Analysis of balance sheets and income statements, as in conventional rating agency methodology;
- Contingent Claims Analysis, an application of Black-Scholes-Merton option theory (the issuer of a default-prone obligation has the option to default and thus “put” the assets of the firm to the bondholders);
- Financial misstatement risk, evaluated by a Big Data approach that compares the balance sheet and income statements of each obligor to comparable financial statements by other obligors, and identifies possible anomalies in financial statements;
- Natural Language Processing (NLP) of commentary from social media and news sources. NLP is a tool that sweeps public sources for changes in sentiment and seeks to capture changes in information in a timely fashion.
- Matrix pricing of infrequently-traded bonds allows investors to identify profit opportunities in the less liquid part of the Chinese corporate bond universe. Greifenberg’s proprietary matrix pricing model identifies reference bonds from the liquid universe, and estimates the fair value of illiquid bonds based on the reference price of comparable liquid market, adjusted for the risk characteristics of illiquid bonds.
Greifenberg’s portfolio system calculates expected return and expected loss for corporate bond portfolios at user-defined confidence levels.