Research On Evergrande’s Financial Risk And Debt Crisis Based On BP Neural Network Model

China’s real estate industry has always been the backbone of national economic development.However,with the deepening of government regulation and the gradual saturation of the real estate market,the crazy development of the real estate market has been curbed.Looking back on 2021,China’s real estate industry has experienced challenges,and many real state enterprises are facing the plight of bankruptcy.Financial disorder,blind expansion,poor operation of capital chain and irregular internal management are the main reasons for the crisis or even bankruptcy of real estate enterprises in China.The introduction of the "three red lines" policy has also exposed the debt crisis of real state enterprises to the spotlight,and the previous mode of high leverage,high investment and rapid expansion of real state enterprises has become history.Corporate financial risk warning is a subject widely studied by scholars all round world.It not only has academic value,but also has great application value.It can be an important reference for company managers,investors,and other stakeholders.Scholars over the world have studied many widely used financial risk warning and evaluation models,but with the development of economy,the financial management and business model characteristics of various industries are not the same,so creating specific industry financial risk early warning models is necessary.According to the background above,this paper selects the debt crisis of Evergrande as the research object.Firstly,based on the review of the relevant literature on corporate financial risk,combining the characteristics of China’s real estate industry,this paper constructs a specific financial risk warning index system for China’s real estate enterprises from seven dimensions and uses the K-W examination and the factor analytic method to carry on the screening and the optimization to the target.The paper obtaines 7 factors of 7dimensions separately,uses the BP neural network model training and the simulation,outputs sample enterprise in three warning degrees financial risk warning result.In this paper,the four fast learning methods of trainlm,trainrp,trainscg and trainbfg in the neural network toolbox of MATLAB are compared,and the model with the best training effect is selected.The simulation test of the model reaches 96% accuracy and has strong recognition ability.Then,this paper goes deep into the case of Evergrande debt crisis,combs the whole story of its debt crisis,and makes a preliminary analysis of its financial indicators,to have a general grasp of its financial situation and business development events.Using the trained model to judge its financial situation in recent years,two financial risk stages from 2015 to 2017 and from 2019 to 2022 are obtained,and in-depth analysis is made on these two stages in combination with Evergrande’s strategic development events.Finally,according to the above analysis and other relevant information,this paper summarizes the causes of Evergrande debt crisis from three aspects and gives relief suggestions.The macro market environment,the characteristics of the real estate industry and Evergrande’s own problems are all factors leading to Evergrande’s debt crisis.In view of these problems,this paper puts forward relevant suggestions: at this stage,the most important task of Evergrande is to slow down the pace of land acquisition,ensure the completion of the delivery plan of all contract sales,and actively reduce liabilities in accordance with the requirements of the "three red lines" to comply with regulatory requirements.Looking forward to the future,Evergrande needs to eliminate diversified projects with poor profitability and little strategic significance,actively seek transformation,optimize internal financial management and corporate governance capabilities,and ensure sustainable development.