A Logistic Regression-based Early Warning System for Financial Risks under the Digital Transformation of Manufacturing Enterprises

With the rapid development of the digital economy, manufacturing enterprises have become the main battlefield for developing the digital economy, and digital transformation is a general trend. Various uncertainties encountered during the process of digital transformation of manufacturing enterprises, how to identify risks in advance in the digital transformation, timely risk avoidance, and risk control are critical to the transformation's success. As a result, establishing a financial early warning system for manufacturing enterprises undergoing digital transformation is critical. This paper introduces non-financial indicators on the basis of financial indicators, takes into account digital transformation, and uses Python crawlers to use text analysis to obtain the digital transformation of 20 manufacturing companies listed on the Shenzhen and Shanghai exchanges that are treated as ST in 2018-2022 and 20 companies in the same industry and of the same scale but with normal finances in that year. The financial indicators were then filtered using the K-S test, t-test, and Mann-Whitney U test, yielding a final set of 18 indicators. The final six principle components were extracted by factor analysis using the KMO and Bartlett tests, and a logistic regression financial risk early warning model was built with them to serve as a guide for manufacturing businesses and their stakeholders going through a digital transition.