Predicting Credit Bond Default with Deep Learning: Evidence from China

China’s credit bond market has rapidly expanded in recent years.However,since 2014,the number of credit bond defaults has been increasing rapidly,posing enormous potential risks to the stability of the financial market.This study proposed a deep learning approach to predict credit bond defaults in the Chinese market.A convolutional neural network(CNN)was selected as the classification model and to reduce the extreme imbalance between defaulted and non-defaulted bonds,and a generative adversarial network(GAN)was used as the oversampling model.Based on 31 financial and 20 non-financial indicators,we collected Wind data on all credit bonds issued and matured or defaulted from 2014 to 2021.The experimental results showed that our GAN+CNN approach had superior predictive performance with an area under the curve(AUC)of 0.9157 and precision of 0.8871 compared to previous research and other commonly used classification models-including the logistic regression,support vector machine,and fully connected neural network models-and oversampling techniques-including the synthetic minority oversampling technique(SMOTE)and Borderline SMOTE model.For one-year predictions,indicators of solvency,capital structure,and fundamental properties of bonds are proved to be the most important indicators.

Microstructure,Mechanical,and Tribological Properties of Amorphous WB2/Ti Multilayer Coatings

Amorphous WB2/Ti multilayer coatings with different modulation ratios(1,5,15,and 30)and different bilayer numbers(10,20,30,50,and 70)were deposited by direct-current magnetron sputtering.The effect of modulation ratio and bilayer number on the microstructure,mechanical,and tribological properties of WB2/Ti multilayer coatings was studied systematically.In the investigation for the influence of modulation ratio,the hardness of the multilayer coatings increases with the increment of modulation ratio.The indentation toughness shows a different changing tendency compared to that of the hardness with different modulation ratios.The wear rates of the multilayer coatings with different modulation ratios are at the same level with the influence of both hardness and indentation toughness.In the investigation for the influence of bilayer number,both hardness and indentation toughness increase first and then decrease with an increasing bilayer number.WB2/Ti multilayer coating with a modulation ratio of 5 and bilayer number of 50 exhibits the highest hardness of 20.4 GPa and lowest wear rate of 1.76×10−16 m3/Nm.In addition,the hardness strengthening mechanism and wear mechanism of the multilayer coatings were discussed in detail.