Empirical Study of the Financial Risk Management Based on Multivariate Statistical Techniques

Financial crisis early warning analysis is important for enterprises, commercial banks and various investors. The aim of my study was to determine if the Logistic multi-regression model enhances the efficiency veracity of financial crisis early warning. In this thesis, the author choose 63 A stock companies, which are marked ST companies because of abnormal financial standing in Shanghai and Shenzhen in 2006, form the financial crisis sample, and choose some similar sized listed companies in same industry as matching sample, Finding index of remarkably distinct the financial crisis company and the non-financial crisis company through analyzing financial indexes of the listed companies. Taking the index of property liabilities ratio, audit opinion, finance lever ratio, gross property net profit ratio, sales revenue growth ratio and cash flux to current liability ratio as the final variants, set up the Logistic multi-regression model, conduct the case analysis of financial crisis early warning. The empirical results show that Logistic regression model help to improve the efficiency of financial crisis early warning.