Research On Financial Crisis Prediction Model Of Listed Companies Based On Ensemble Learning Algorithm

With the rapid development of China's capital market,enterprises are facing increasingly fierce competition.Because of the financial crisis,there are 321 listed companies in China A-stock had got special treatment from 2006 to 2015.This not only made the companies suffer from huge losses,but also had significantly bad influence on stakeholders and even the market.Therefore it has important implications that we use statistical methods to establish a reliable and stable financial prediction model.If we can predict the likelihood of financial distress correctly,that will be practical significance for protecting the interests of investors and creditors,Operators' preventing the financial crisis,Government departments' monitoring quality of listed companies and stock market risk.Initiated by background introduction and potential contribution,this paper establishs the research framework and potential innovative ideas based on current study.This paper then systematically reviews the literature research by both domestic and foreign scholars in the area of enterprise financial crisis on aspects of definition,cause of formation,index systems and model building particularly.Based on these,the paper clarifies the definition of financial crises and also the primary approach to index selection from an empirical point of view.This paper uses ensemble learning model to determine whether the listed companies will suffer the financial distress or not.This paper selects the listed companies which have got special treatment from 2006 to 2015 as the research object.By comparing with the companies which have the same scale and the same industry,this papeer studies the main characteristics of the listed companies,including the corporate governance and the financial statements.So this paper evaluates the solvency,operating capacity,profitability,cash flow and potential development of listed companies,which can help us having a clear understanding of the listed companies suffered financial distress.Based on a certain understanding of characteristics of the listed companies suffered financial distress,we choose eleven prediction variables to establish a reliable and stable financial crisis prediction model using the randomForest algorithm.When using the test sample from 2014 to 2015,the accuracy rate is 83%,which is 4 to 20 percentage points higher than the traditional classification method.Meanwhile,the model is used in different parameter,and it is proved to work well,showing a good adaptability and stability.Finally,this paper find that the EPS and ROE indicators play a very important role in distinguishing the company suffered financial distress.Because of the uncertain factors,the article still has some shortcomings,such as the limited sample size which can not cover all the listed companies,and the selected indicators which also fail to reflect the cause of getting the listed companies in financial distress.Based on these findings,the ensemble learning model is very useful to predict the company suffered financial distress,especially for managers,investors and creditors...