The framework of Hammer (CAFÉ) credit rating for capital markets in China with international credit rating standards

The goal of this paper is to discuss how we establish the so‐called “Hammer (CAFÉ) Credit System” by applying the Gibbs sampling algorithm under the framework of a big data approach supported by both traditional structured and unstructured data as a breakthrough, in particular, to extract those highly related risk features in depicting default (bad) events and related fraudulent behaviors (action) by following the “five step principle” incorporated with the international credit rating standards in the practice. The analysis shows that our Hammer (CAFÉ) Credit System is able to handle current three issues raised by the credit rating business for capital markets in China, which are as follows: (1) The rating is falsely high, (2) the differentiation of credit rating grades is insufficient, and 3) the performance in predicting early warning and related issues is poor. In addition, the Hammer (CAFÉ) credit discussed in this paper is supported by clearly defining the “BBB” grade rate as the basic investment level associated with the annualized rate of default probability, and the credit transition matrix for “AAA‐A” to “CCC‐C” credit grades in accordance with international standards used in the practice of risk management and decision services.