Analysis of corporate financial risk avoidance strategies based on logistic regression model

Abstract

Abstract The gradual accumulation of financial problems of enterprises will form financial risks. If financial risks are identified and solved in time, the losses brought by financial risks to enterprises can be reduced, and even the formation of financial crises can be avoided. This paper first explores the limitations of different forecasting methods and conducts in-depth exploration and research on enterprise financial risk management from risk management theory and enterprise cycle theory. Secondly, we analyze the application of GBDT and logistic regression models in financial risk early warning management, elaborate on the idea of combining GBDT and logistic regression, and construct a combined financial risk early warning model based on the combination of GBDT and logistic regression. Finally, seven dimensions and 34 indicators are used to measure the financial risk prediction ability to make prediction analysis and model evaluation of the financial risk of listed companies. The results show that the risk prediction accuracy of the combined model of GBDT and logistic regression is 91.25%, which is significantly higher than that of the single model of logistic regression, proving the effectiveness of the combined model for financial risk early warning. This study establishes an effective financial risk early warning model to provide scientific references and suggestions for managers and investors of listed companies.