Research On Warning Of Financing Risk Of Listed Companies In Digital Product Manufacturing Industry Based On LSTM

Since 《The 14 th five year plan for national economic and social development of the People’s Republic of China and the outline of long-term objectives for 2035》proposed that it is necessary to deeply implement the strategy of manufacturing a strong country and promote the high-quality development of the manufacturing industry.It is necessary to implement the transformation and upgrading and high-end intelligentization of the manufacturing industry.At the same time,it also proposed the goal of "accelerating the promotion of digital industrialization" and improving the level of related industries.This is an important strategic guidance made by the Party Central Committee based on the current industrial situation and industrial development trend of the manufacturing industry.As an important key link in realizing the industrialization of data elements,digital product manufacturing plays an important role in the integrated development of digital technology and the real economy.Therefore,the rapid and stable development of the digital product manufacturing industry has a significant impact on the development of my country’s social productive forces and the transformation of production relations.Most of the products produced by digital product manufacturing companies are high-tech products,which have high technical difficulties.Therefore,greater innovation and investment are needed to improve the manufacturing level of enterprises.Therefore,the problems of high capital risk,high income uncertainty and long cycle are very prominent.How to deal with the slowdown of economic growth,reduce the willingness of consumption,reduce the financial pressure caused by the sudden attack of the superimposed epidemic,early warning and positioning the financing risks faced by enterprises,alleviate the financial difficulties and financing pressure faced by digital product manufacturing industry,realize the development of digital product manufacturing industry,industrialize and market digital elements,and realize the integrated development of digital technology and real economy is an important problem to be solved urgently.With the deep development of economic globalization and the complex external situation brought by the impact of the epidemic,the factors affecting the financing risk of digital product manufacturing enterprises also show a complex and changeable situation.It includes not only the factors affecting the financing mismatch within the enterprise,but also the external factors such as policy environment and industry environment.The existing research perspective on the impact of financing risk of digital product manufacturing industry is too scattered and has less content.It is not able to investigate the financing risk of digital product manufacturing enterprises in combination with internal and external common factors.At the same time,the traditional financial risk early warning model can not effectively identify the impact of the superposition of multiple indicators,and there are missing variables,resulting in a significant reduction in the accuracy of early warning.Therefore,this thesis introduces web crawler,text analysis and grounded analysis,long short term memory(LSTM)model and xgboost model to construct the financing risk early warning system of Listed Companies in digital product manufacturing industry.This thesis takes the risk management theory as the core,and carries out the early warning of the financing risk of the digital product manufacturing industry from the three dimensions of the financing industry environmental risk,the financing term mismatch risk and the financing structure mismatch risk.The analysis is carried out from the early warning process of "risk intelligent identification-risk intelligent early warning-risk early warning positioning".The main research contents of this thesis include:(1)The intelligent identification of financing risks of listed companies in the digital product manufacturing industry is realized.By using web crawler technology to obtain the financing risk literature of listed companies in the digital product manufacturing industry,using text analysis to intelligently analyze risk keywords,and using grounded analysis methods to analyze digital product manufacturing from three dimensions: financing term mismatch,financing structure mismatch,and external environment.Based on the identification of financing risk factors of listed companies in the industry,a financing risk early warning indicator system is constructed under the guidance of risk management accounting thought.(2)Build an intelligent early warning model for financing risk of Listed Companies in digital product manufacturing industry.By constructing the intelligent early warning model of financing risk based on LSTM,and through the training,performance test and robustness test of LSTM model,the accuracy of LSTM in the field of financing risk early warning of Listed Companies in digital product manufacturing industry is confirmed.The intelligent early warning monitoring of financing risk of Listed Companies in digital product manufacturing industry is realized,and the financing risk alarm degree of enterprises in 2022 is predicted.(3)Realize the early warning positioning of financing risk of Listed Companies in digital product manufacturing industry.By using the characteristic importance score of xgboost model to sort the financing risk indicators of Listed Companies in digital product manufacturing industry,identify the indicators that have the greatest impact on financing risk,and carry out targeted prevention and resolution of the risk sources represented by the key indicators of high risk,so as to realize the design of prevention and control measures for financing risk.This thesis tentatively studies the financing risk early warning of Listed Companies in digital product manufacturing industry from the perspective of "financing external environment-financing term mismatch-financing structure mismatch",and selects targeted preventive measures according to the early warning positioning results,which has a certain practical value.This thesis attempts to innovate in the following three aspects:(1)apply intelligent technologies such as web crawler and text analysis to the intelligent identification of financing risks of Listed Companies in digital product manufacturing industry;(2)The LSTM model is introduced into the field of financing risk early warning,which improves its accuracy compared with the traditional early warning model;(3)The characteristic importance score of xgboost model is used to locate the index importance of financing risk of Listed Companies in digital product manufacturing industry,accurately locate the alarm source of enterprise financing risk,and provide the basis for policy formulation and the implementation of prevention and control measures.