A Study Of The Steel Industry Listed Companies’ Financial Risk Prediction Based On BP Neural Network

Iron and steel industry is the pillar industry of the national economy, now the globaleconomy continues to slow and iron ore and energy are so high prices, the operatingperformance of China’s iron and steel industry which was already in the stage of excessproduction capacity deteriorates, profitability has fallen sharply, and the financial risks.The steel industry in the event of financial risk, not only endangers their own survival anddevelopment, also will bring losses to investors and other related industry. Therefore, tobuild an effective and practical steel industry listed companies’ financial risk predictionmodel which meets the needs of stakeholders is increasingly urgent, and that has greatpractical significance.This article is to proceed from a theoretical point of view of intelligence, first of allrough set theory and its relative attribute reduction and the basic principle of BP neuralnetwork are presented, based on the above theory, proposes a rough set and BP neuralnetwork technology method of combining, this method is applied to China’s steel industrylisted companies’ financial risk early warning research.First, describes China’s steel industry listed companies’ financial risk early warningresearch background, significance and research status, and points out the previous researchresults and practicability, demonstrates the necessity of the study.Secondly, financial risks have been defined, and a detailed analysis of its form factorsare made, and the detailed analysis of the characteristics of the complementary strengthsof a combination of rough set theory and BP neural network technology, provides a basisfor the selection of the early warning indicators in the next part;Thirdly, introduces the forms of our country’s steel industry listed company financialrisk and does the comprehensive analysis of external and internal factors which affect therisks, on the basis of the analysis combined with the characteristics the iron and steelindustry to select the financial and non-financial indicators which can show steelcompanies’ financial position to build the financial risk early warning indicator system;Finally, selected the30listed companies in the steel industry as the research sample,the indicator data of the sample was processed in accordance with the method described above. Because of the limitations of traditional methods for early warning model, thisartical creatively used hierarchical cluster analysis to divide the the sample enterprises’financial situation into a five-level progressive, and constructed of BP neural networkprediction model, multi-level classification of financial position for the early warningmodel provided accurate output layer target. After training of BP neural network, used testsamples to test it and proved that the model works well. Experimental results confirmedaccording to the steel industry listed companies by building of rough set, the BP neuralnetwork financial early warning model is effective.