大数据驱动企业财务风险预警的有效性研究——以A股上市公司为例

The Effectiveness of Big Data-Driven Financial Risk Early Warning for Enterprises—Evidence from A-Share Listed Companies

摘要: 大数据技术在企业财务风险预警中的应用日益广泛，基于A股上市公司2018~2022年的财务数据和非财务数据，构建了包含财务指标，市场指标和文本数据的多维度预警指标体系。研究采用机器学习方法，通过对比传统财务指标模型与大数据驱动模型的预警准确率，验证了大数据在提升企业财务风险预警效果方面的积极作用。实证结果表明，大数据驱动的预警模型在识别财务困境企业方面的准确率显著高于传统模型，且能较早发出风险预警信号。

Abstract: The application of big data technology in corporate financial risk early warning has become increasingly widespread. Based on financial and non-financial data of A-share listed companies from 2018 to 2022, this study constructs a multi-dimensional early warning indicator system incorporating financial indicators, market indicators, and textual data. Using machine learning methods, the study compares the warning accuracy between traditional financial indicator models and big data-driven models, validating the positive role of big data in enhancing corporate financial risk early warning effectiveness. Empirical results demonstrate that the big data-driven warning model significantly outperforms traditional models in identifying financially distressed companies and can issue risk warning signals earlier.