Credit Rating of Chinese Companies Based on XGBoost Model

With the outbreak of the COVID-19 epidemic, the global economy is on the downswing and the credit crisis is coming. In order to prevent credit risk and further standardize credit rating methods, this paper innovatively introduces the machine learning method-XGBoost model to credit rating based on financial indicator data of 1021 listed Chinese companies in 2020 and real bond default data in 2021. By comparing with the logistic regression model, it is found that the XGBoost model has better prediction effect, and its output index importance score can provide guidance for enterprises to manage their own credit ratings.