Research on enterprise financial crisis early warning management based on PLS-BP

Abstract

Finding an efficient and accurate early warning method is of great significance to enterprises, investors and the development of China's national economy. The Partial Least Squares (PLS) method is used to screen the variables, the BP neural network is used to establish the model and the PLS-BP early warning model is constructed. In the simulation experiment, the prediction accuracy of PLS-BP early warning model in T-2 period is 100%, 93.75% in T-3 period and 87.5% in T-4 period, which are significantly higher than the traditional financial early warning methods. The above results prove that PLS-BP early warning model can accurately predict the financial crisis of enterprises and has certain practical value.