Research on financial early warning of listed companies based on factor analysis and logistic regression

Abstract

The establishment of an accurate, effective, sensitive and stable financial crisis early warning model for listed companies has a role that cannot be underestimated for the stability and healthy development of China's financial order. In this paper, financial statements of listed companies in Shanghai and Shenzhen Stock markets from 2018 to 2021 were selected as sample data. On the basis of the four traditional financial indicators, namely profitability, solvency, operating ability and development ability, two new financial indicators, namely profit quality and market valuation, were added. The factor analysis method and binary Logistic regression method are combined to construct the financial early warning model of listed companies. The results show that the model has high fitting accuracy and discriminant ability, which proves that the financial indicators disclosed by listed companies can provide empirical basis for various stakeholders to provide an empirical basis for early warning of corporate financial risks.