The effect of green, supply chain factors in predicting China’s stock price crash risk: evidence from random forest model

Exploring the essential factors to prevent the risk of a stock price crash is a worthy issue. The occasional supply chain interruption significantly impacts stock price fluctuation and even leads to a crash. This paper analyzes the effect of green and supply chain factors on stock price crash risk based on interpretable machine learning. A new research framework of data-driven theory is presented, which overcomes the shortcomings of traditional econometrical methods. The results show that the random forest regression model established has higher accuracy and lower misclassification rate in predicting and identifying the crash risk. Furthermore, this paper introduces the interpretable machine learning method to disassemble the black box of stock price crash risk prediction, makes use of the feature importance method, partial dependence plot, and employs the Shapley value. The analysis based on Chinese enterprises shows that environmental improvement can decrease stock price crash risk, and supply chain factors are important in affecting the crash risk. Liquidity and leverage have the most significant impact on the risk of stock price crash of polluting firms, and supply chain factors are more influential in predicting share price crash risk among non-SOEs.