Application of BP neural network technology on dynamic financial risk prediction

Abstract

This work mainly studied the application of BP neural network technology. Based on the crisis early warning of listed real estate companies at home and abroad, this study used BP neural network method to determine the corporate governance indicators of crisis enterprises from eight aspects, such as profitability, solvency, operation ability, development ability, cash flow, risk, etc. This work took the real estate listed companies as the research object and put forward eight early warning indicators. Traditional early warning models have low fault tolerance, and most of them are static early warning. In this study, the innovation point is that the BP neural network model has the ability of self-learning and adjustment, strong risk identification and fault tolerance, and can cope with the changeable financial risk environment of real estate enterprises. BP neural network model can analyze problems through self-learning and training, and has strong nonlinear mapping ability. It is feasible to construct financial risk early warning model of real estate listed enterprises.