Early Warning Of The Financial Risk Study Based On Logistic Regression Models

Abstract：Since2008，the financial tsunami triggered by the U.S. subprime mortgagecrisis and the EU sovereign debt crisis continued to simmer. A large number ofenterprises have experienced sustained loss and insolvency, while the world economyremains depressed, exhibiting a stagnant scenario and sluggish growth. In order tosurvive the cruel competition in this complex the business environment, enterprises inour country must improve the quality of their financial management and implementeffective methods for early-warning of financial risk.The purposes of this study is to learn from previous results, to use statistical teststo choose and analyze various indicators, which reflect a company’s financialsituations and potentials, to apply Logistic-regression method to create financial riskearly-warning models for listed companies of certain specific fields in our country,and to help these companies to find out and defuse the crisis in an early stage，so thatthese companies can embark on the track of sustained and healthy business.The main contents are arranged as the following. First, I started with a briefintroduction to the research background. Second, I reviewed the current researchresults of applying Logistic-regression method to create financial risk early warningmodels and compared the pros and cons of the current situations of this study in ourcountry and overseas, in order to explain the goals, significance, ideas and methods ofthis study. I also discussed the application premise, scope, advantages, disadvantagesand feasibility of the Logistic-regression-based model, and laid a theoreticalfoundation for applying this model in the following chapters. Third, twenty fourindicators are chosen from the following four aspects of a corporation, includingrevenue, solvency, operational capacity, development capacity, to build a system forevaluating the early-warning financial risks of corporations. Ninety listed companiesin machinery, equipment, and instrument industry are then analyzed for buildingearly-warning models using the Logistic regression method. Fourth, using TB Co. Ltd, as an example, I apply the model to its financial data of2011for making predications.In accordance to the results, comments and suggestions are made to improve thecompany’s internal management to lower its financial risk. At last, future applicationprospects of the model are also discussed.