Application of logistic regression model in corporate financial decision making in the era of big data

Abstract

In order to improve the prediction accuracy of financial decision risk early warning model (FEW), this paper firstly introduces five types of non-financial indicators, market price, management level, corporate reputation, governance structure and audit index, into the logistic regression (Logistic) financial risk decision early warning model on the basis of traditional financial indicators. Secondly, using the sample of A-share ST companies and paired non-ST companies in 2021, the probability of a company being ST after 3 years was predicted with 2018 data, and the robustness of the selected logistic financial risk decision early warning model was further tested by comparing the early warning accuracy of six logistic financial risk decision early warning models. The results of the study showed that adding three indicators, namely, market price, management level, and corporate reputation, to the logistic model was effective in improving the early warning accuracy of the model by 23.6%. In addition, by comparing the selected logistic financial risk early warning model with the Z-score model, it is found that the logistic model has the highest decision warning accuracy of 95.3%, which has a high robustness. Therefore, this paper introduces the logistic regression financial risk decision early warning model effectively helps enterprises to make financial decisions to issue timely warnings before the occurrence of unbearable crises and take appropriate measures in advance to avoid the enterprise delisting crisis.